The membrane potential

In the dark, rods have a resting membrane potential of about −40 mV. The negative value means that there is more positive charge on the outside of the plasma membrane than on the inside. In most resting neurons, the potential is closer to −70 mV, so rods are considered to be relatively depolarized as compared to other cells or, as we shall see, as compared to fully light-activated rods. In rods, as in other neurons, the source of membrane potential is the inequality of ion concentrations on either side of the plasma membrane generated by the Na ⁺ /K ⁺ -ATPase. This protein uses the energy released upon ATP hydrolysis to pump sodium ions out of the cell and potassium ions into the cell, and in rods it is found in the plasma membrane of the inner segment (see Chapter 19 ). In most neurons, potassium channels are the predominant conductance at resting potentials, and these allow potassium ions to flow out of the cell down their concentration gradient. Because these channels pass a specific cation, but no anions, a net positive charge accumulates on the outside of the membrane and a net negative charge accumulates on the inside of the membrane. This occurs until the electrical energy stored in the charge separation equals the energy stored in the potassium gradient, the relationship described by the Nernst equation, described in detail in many textbooks. Strictly speaking, this equation only works when the membrane in question contains a single type of channel permeable to only a single type of ion. In reality there are always additional minor conductances (other channels and transporters) so that the resting potential is modified by these to give the typical resting value of about −70 mV to −90 mV for excitable cells.

The dark current and the cGMP-gated channel

Rods also have a negative resting membrane potential, but its magnitude is less than in many other neurons because the effects of potassium channels in the inner segment are balanced by a major additional channel current in the outer segment. The less negative resting potential, or partially depolarized state, of rods is due to the cyclic-GMP-gated cation channel (see Chapter 19 ), also known as CNG channel (for cyclic nucleotide-gated) or light-sensitive channel or phototransduction channel (see Chapter 19 ). This multi-subunit channel only passes cations, but does not discriminate strictly among different cations, allowing Na ⁺ , K ⁺ , and Ca ²⁺ to pass. Among physiologically important cations it has the highest permeability to Ca ²⁺ , but because of its much higher concentration outside the cell (over 140 mM, as compared to ~1.5 mM for Ca ²⁺ ), Na ⁺ is the primary carrier of the current through the CNG channel in the dark. Because it tends to dissipate the gradient of charge, by allowing positive charge to enter the cell, the open CNG channel causes the cells to be partially depolarized. Put another way, because there are substantial inward sodium currents in the outer segment to balance outward potassium currents in the inner segment, the resting potential is moved away from the potassium “equilibrium” or Nernst potential of ~−90 mV, and toward the sodium “equilibrium” or Nernst potential of ~+60 mV. Because there is net flux of cations out of the inner segment plasma membrane, and a net flux of cations into the outer segment plasma membrane, as well as electrical conductance between the inner and outer segments, a complete circuit is made which is known as the circulating current or dark current.

Ca ²⁺ and the exchanger

The CNG channel also has an influence on the concentration of Ca ²⁺ , which in rods and cones, as in other cells, plays an important role in signaling and regulation. A Na ⁺ /Ca ²⁺ , K ⁺ exchanger protein, NCKX2, uses the energy stored in the Na ⁺ gradient across the plasma membrane to push Ca ²⁺ ions outside the cell, against their concentration gradient (see Chapter 19 ). In the absence of any inward flux of Ca ²⁺ , this mechanism would lower the Ca ²⁺ concentration in the outer segment cytoplasm to about 10 nM. In the dark, there is an inward leak of Ca ²⁺ through the CNG channel, so the resting level of Ca ²⁺ is a few hundred nanomolar.

Even though in the dark more channels are open than at any other time, because cGMP levels are at their highest, the percentage of total channels that are open is fairly small; most of the channels are closed even in the dark (see Chapter 19 ). One reason for this situation is that binding of more than one cGMP is necessary to give each channel a high probability of opening, i.e. the response of the channel to cGMP is non-linear and shows positive cooperativity. The other reason is that the dark concentration of cGMP is well below the concentration of cGMP at which channel opening probability is 50 percent, a number that reflects the affinity of cGMP for the channel subunits as well as the positive cooperativity. Because the dark cGMP concentration is at the low end of the dose response curve for the channel (therefore very far from saturation), even a small change in cytoplasmic concentration of cGMP can be immediately sensed as a change in the number of open channels and therefore of the dark current.

Control of [cGMP] by guanylate cyclase and PDE6

The resting, or dark, level of cGMP is determined by the balance between the activity of the enzyme that synthesizes it, guanylate cyclase (GC), and the enzyme that degrades it, the cGMP phosphodiesterase, PDE6. Both of these have much lower activities in the dark than they do in the light. When fully activated PDE6 is one of the most efficient enzymes known, and the most active of any of the PDE superfamily of cyclic nucleotide phosphodiesterases. The maximal turnover at saturating cGMP concentrations, k _cat , has been reported to be between 1000 s ⁻¹ and 7000 s ⁻¹ , and the Michaelis constant, the cGMP concentration at which activity is half-maximal is about 40 µM. Thus the enzymatic efficiency, k _cat / K _m , is thus in the range of 2.5 × 10 ⁷ M ⁻¹ s ⁻¹ to 1.75 × 10 ⁸ M ⁻¹ s ⁻¹ , implying that it operates near the diffusion limit. PDE6 requires two kinds of metal ions for activity. Mg ²⁺ binds reversibly along with cGMP, and Zn ²⁺ is permanently bound to high affinity sites essential for catalysis. The other key players in photoactivation are also in quiescent states under dark conditions. The mechanisms for regulation of these enzymes by light are discussed below.

Guanylate cyclase uses GTP as a substrate, and produces cGMP and inorganic pyrophosphate as products. Inorganic pyrophosphate is rapidly broken down by the enzyme inorganic pyrophosphatase, which is found at higher levels in rod outer segments than in any other cell type in which it has been measured. Guanylate cyclase is a transmembrane protein that is thought to exist in photoreceptor membranes as a dimer, with each dimer bound to the calcium binding protein, GCAP (guanylate cyclase activating protein). PDE6 is a heterotetrameric peripheral membrane protein consisting of two large catalytic subunits, PDE6α and PDE6β, as well as two smaller identical inhibitory PDE6γ subunits. The catalytic subunits are each subject to the set of post-translational modifications associated with isoprenylation, as described below, so the enzyme is bound in a peripheral way to disk membranes. The PDE6α and PDE6β subunits bound in rods are similar in structure to one another, whereas in cones, two identical PDE6α subunits are bound. PDE6 catalytic subunits are related in sequence to a large family of cyclic nucleotide phosphodiesterases, the PDE superfamily. These share sequence similarity in their catalytic domains, where the PDE6 isoforms most closely resemble in structure the PDE5 isoforms, which are the targets of sildenafil citrate (Viagra) and other drugs used to treat erectile dysfunction. In addition to the catalytic domains, PDE6 catalytic subunits each contain two GAF domains, one of which contains a non-catalytic binding site for cGMP. Because the total concentration of cGMP-binding GAF domains in the outer segments is about 50 µM, most of the cGMP in the cell is bound to PDE6. The physiological function of these non-catalytic sites is unclear, but their occupancy is coupled to binding of the catalytic subunits by the inhibitory PDE6γ subunit. It is the PDE6γ subunit which interacts most extensively with the activated form of the G-protein, as described below.

Rhodopsin

In the dark, the photon receptor rhodopsin is in its inactive state with an inverse agonist, 11- cis retinal, covalently attached to its active site ( Figs 18.5–18.7 ). Ligands that activate receptors are called agonists, those that block the action of agonists are called antagonists, and those that actually reduce receptor activity below the level it has in the absence of any ligand are called inverse agonists. Rhodopsin has seven membrane-spanning helices and is a part of the G protein-coupled receptor (GPCR) super-family of signal transducing receptors. Like other GPCRs, its activity consists of its ability to catalyze the activation of a heterotrimeric G-protein. This is the activity of rhodopsin which is inhibited by the chromophore and inverse agonist 11- cis -retinal. The apo-form of rhodopsin without a ligand, known as opsin, has much higher G protein-activating activity than rhodopsin does, a fact that becomes important at high light levels but is relatively unimportant in the dark-adapted state. Rhodopsin is present at very high concentrations in the disk membrane (and somewhat lower concentrations in the plasma membrane), so that about one-third the surface area of the disks is occupied by rhodopsin, and the other two-thirds is occupied by phospholipids, cholesterol, and other minor lipids.

Surface rendering of rhodopsin crystal structure with cut-out of retinal binding pocket (red) to show 11- cis retinal chromophore (yellow spacefill). PDB coordinates, 1U19.

Crystal structures highlighting helical arrangement in lipid bilayer of opsin (left, PDB structure 3CAP ), dark rhodopsin (middle, PDB structure 1U19 ), and active opsin with peptide derived from the alpha subunit of transducin (right, PDB structure 3DQB ).

( A ) Molecular structure, in space-filling representation, of 11- cis retinal in rhodopsin. The lower left depicts the sidechain of lysine 296, with the Schiff’s base nitrogen shown in blue. Carbon atoms are shown in red, and hydrogens in white. ( B ) Molecular structure, in space-filling representation, of all trans retinal in rhodopsin. The lower left depicts the sidechain of lysine 296, with the Schiff’s base nitrogen shown in blue. Carbon atoms are shown in yellow, and hydrogens in white.

Rhodopsin is subject to post-translational modifications that are important for its function. N-linked carbohydrates are found on the domain of rhodopsin that faces the inside of the disks and the extracellular solution, and they appear to be important for proper transport of rhodopsin from its site of synthesis in the inner segment to the outer segment membranes. Two palmitoyl groups are attached via thio-ester linkages to two adjacent cysteine residues near its carboxyl terminus, which tether the polypeptide chain to the cytoplasmic surface of the disk membrane forming a fourth cytoplasmic loop. In addition, the aldehyde moiety of 11- cis retinal and all- trans retinal is not bound to rhodopsin as a free aldehyde, but rather a covalent Schiff’s base linkage is formed between ll- cis retinal and lysine residue 296 in the transmembrane domain of rhodopsin ( Fig. 18.7 ).

There are 348 amino acids that comprise rhodopsin. Whereas mutations at only four different positions are found in patients with the disease congenital stationary night blindness (CSNB; Box 18.1 ), there are over 100 amino acid positions that are found mutated in patients with the blinding disease autosomal dominant retinitis pigmentosa (ADRP; Box 18.2 ). Therefore, the molecule is not only sensitive in that it can become activated with only one photon of light, it is also remarkably sensitive to its amino acid composition.

Box 18.1

Congenital stationary night blindness

Congenital stationary night blindness (CSNB) comprises a group of genetically and clinically heterogeneous non-progressive retinal disorders, mainly due to defects in rod photoreceptor signal transduction and transmission. Mutations in genes coding for proteins of the phototransduction cascade ( RHO , GNAT1 , PDE6B , RHOK , SAG and RDH5 ) or genes associated with the transmission of the signals from the photoreceptors to bipolar cells ( NYX , GRM6 , CANCA1F and CABP4 ) can lead to this disease.

Night blindness, reduced or absent dark adaptation, is a typical and early sign of various forms of retinal dystrophies. Forms differ in inheritance pattern (autosomal dominant, autosomal recessive or X-linked), electroretinograms (presence or absence of a-wave), refractive error (presence or absence of myopia) and fundus appearance ( Fig. 18.8 ). All patients with stationary night blindness have severely reduced rod ERG amplitudes and many have modestly reduced cone ERG amplitudes. Rod sensitivity in patients is decreased by 100× to 1000× compared with normals. Almost all have cone responses with a normal peak implicit time (time interval between the light flash and subsequent peak of b-wave).

Electroretinograms (ERGs) of a patient with CNSB due to rhodopsin A295V show altered signaling. The ERG represents the massed potential changes across the retina in response to light. The early negative deflection is the a-wave, derived from hyperpolarization of the photoreceptors. The later positive deflection is the b-wave, largely derived from depolarization of ON bipolar cells, which receive inputs from the photoreceptors. Scotopic ERGs of a representative control subject (left) and a patient (right). Intensities represented are 1, 0.01; 2, 0.03; 3, 0.03; 4, 1.0; 5, 3.0 cds s m ⁻² . The rod response is severely diminished in the patient.

(Reproduced from Zeitz et al 2008. Reproduced with permission from the Association for Research in Vision and Ophthalmology.)

G-protein, G _t

Like rhodopsin, the G-protein, known as transducin or G _t ( Fig. 18.9 ), is also found in an inactive state in the dark. For a G-protein, this means that it has GDP, rather than GTP, bound to its α subunit, G _tα , and exists in a heterotrimeric form, G _tαβγ . The G-protein associates with the disk membranes by virtue of its covalently attached lipids. ^, A heterogeneous mixture of saturated and unsaturated 12- and 14-carbon fatty acids are found attached to the amino terminal glycine residue of G _tα , linked through an amide bond. This reaction occurs during translation and is important for G-protein localization in the outer segment. There is a 15-carbon isoprenyl group, farnesyl, attached to the carboxyl terminus of the γ subunit ^, through a thioether linkage. This sort of isoprenylation is found in other phototransduction proteins; in each case it occurs at a cysteine residue four residues from the carboxyl terminus of the initial translation product. The final three residues are cleaved by the action of a protease, and the hydrophobicity conferred by the isoprenyl group is enhanced by the methyl esterification of the carboxyl group on what has been converted into a C-terminal cysteine residue.

Structural models for interactions among transducin subunits, metarhodopsin II, and the disk membrane during light activation.

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

(Modifed from Wang et al 2008 .)

Importance of lipid milieu

The lipids of the disk membrane play an important role in assembling the phototransduction machinery and providing the proteins a stage on which to act. Rhodopsin and guanylate cyclase are transmembrane proteins, and G _t and PDE6 are peripheral membrane proteins, attached to the membrane by covalently attached lipids ( Figs 18.9, 18.10 ). GCAPs are also covalently lipidated by fatty acids on their amino termini, and another phototransduction protein, rhodopsin kinase, is isoprenylated (discussed below). The transmembrane CNG channel is found only in the plasma membrane, not in the disks. Experiments with purified proteins have shown that the proper assembly of these proteins on the disk membrane is essential for efficient interactions among them. Rhodopsin diffuses rather rapidly within the lipids of the disk membrane, so that many encounters between it and the G-protein occur every second even in the dark. ROS membranes have an unusual lipid composition, with a high content of polyunsaturated fatty acids. Close to 40 percent of the fatty acyl groups of ROS phospholipids are the ω-3 fatty acid, docosahexaenoic acid (DHA or 22:6), which has 22 carbons and 6 double bonds. DHA can only be synthesized in the body if essential ω-3 fatty acids are consumed in the diet. Cholesterol, while making up only 10–15 percent of total outer segment lipids, plays an important role in establishing the lipid environment in which phototransduction reactions occur.

A multi-subunit complex essential for normal photoresponse recovery kinetics. PDE6γ C-terminal fragment is shown in red space-filling representation.

(Modified from Wensel 2008. )

Photoisomerization of rhodopsin

Phototransduction begins with absorption of light by rhodopsin ( Figs 18.9, 18.11 ). Rhodopsin has a broad absorption spectrum with a wavelength of peak absorbance close to 500 nm in the green portion of the visible spectrum (see Fig. 18.14 below). The probability of capture by rhodopsin of a photon of green light passing through the retina is fairly high (~50 percent for 20 µm long human rods, and ~89 percent for 60 µm long toad rods), due to the large number of disks (~1000) through which it must pass, and the high concentration of rhodopsin in each one. When rhodopsin absorbs light, the probability that it will undergo a structural transition, or photoisomerization, is extraordinarily high: 65 percent. This is one of the most efficient photochemical reactions known, and this efficiency helps explain the exquisite sensitivity of rod cells to a single photon of light. The bound 11- cis retinal chromophore is responsible for the light absorption (polypeptides do not absorb visible light unless they are bound to some chromophoric ligand), and in its excited state induced by the light absorption, 11- cis retinal undergoes a photoisomerization at the bond between carbon 11 and carbon 12 in its structure from 11- cis to all- trans retinal ( Fig. 18.7 ). The rearrangement of bonds causes a dramatic change in the structure of retinal, with the new all- trans form representing a straightening out of a kinked molecule. This molecule no longer acts as an inverse agonist, but now as an agonist, or activator.

Schematic representation of phototransduction activation cascade.

The retinal chromophore, covalently attached via a Schiff’s base linkage to a lysine residue at amino acid position 296 on rhodopsin (see above), induces structural changes in the protein upon photoisomerization. After photoisomerization of the chromophore, the protein undergoes a series of conformational changes in the arrangement of its helices to accommodate this event. In other words, when retinal stretches out into the all- trans shape, it pushes part of the protein out of the way. These conformational changes ultimately lead to formation of Metarhodopsin II, a significantly different structure to dark rhodopsin. The intermediate conformations of photolyzed rhodopsin can be monitored by high speed and/or low temperature spectroscopy , including UV-visible absorbance, infrared, Raman, and fluorescence spectroscopies. These and other studies have shown that upon light activation, rhodopsin first changes to bathorhodopsin, lumirhodopsin, metarhodopsin I (MI) and metarhodopsin II (MII or R*, the form that activates transducin), and ultimately metarhodopsin III (MIII) before the all- trans chromophore becomes hydrolyzed, leaving the apoprotein opsin. At physiological temperatures, almost immediately upon photoactivation, MI and MII establish a dynamic equilibrium which can be monitored by the differences in their absorbance spectra. It is MII rather than MI that preferentially binds to and activates the G protein.

G-protein activation

The key functional difference between rhodopsin and metarhodopsin II is in their interactions with the G-protein transducin (G _t ). MII binds to the G-protein much more tightly than does rhodopsin and serves as an efficient catalyst for the activation of G _t . G-protein activation occurs about 10 million times faster at high MII concentrations than it does in the dark. The G-protein binds to MII with GDP bound, but rapidly releases the bound GDP ( Fig. 18.9 ). This GDP release is the slow step (~once per 10,000 s) for G-protein activation in the dark, but occurs very rapidly when G _t is bound to MII. Once GDP is released, the millimolar concentrations of GTP in the cell ensure rapid binding of GTP. Once GTP is bound to the G _tα subunit, the G-protein is in its active state. The activated G _tα -GTP complex has much lower affinity for both the G _βγ subunits and for MII than do either the nucleotide-free state or the GDP-bound state of G _tα . One MII molecule can activate many G-proteins in a catalytic fashion, at a rate exceeding 150 per second in mammalian rods.

PDE6 activation

When G _tα loses its affinity for MII and G _tβγ upon binding GTP, it also gains greatly enhanced affinity for the cGMP phosphodiesterase, PDE6. ^, When G _tα -GTP is bound, the catalytic activity of PDE6 goes up about one-thousand fold. This dramatic increase in enzymatic activity means that cytoplasmic levels of cGMP begin to fall rapidly in the vicinity of activated PDE6. Fully activated PDE6 is one of the most efficient enzymes known, with a catalytic efficiency, k _cat /K _m , of about 4 × 10 ⁸ M ⁻¹ -s ⁻¹ . This high catalytic efficiency means that nearly every time a cGMP molecule collides by diffusion with an activated PDE6 molecule, it will be hydrolyzed to form 5′-GMP. If 350 PDE6 molecules are fully activated in a single primate rod cell, they can hydrolyze most of its cGMP in one-tenth of a second.

Channel closing

Individual cGMP molecules are constantly dissociating from and binding to their sites on the CNG channel, even in the dark (see Chapter 19 ). Thus each CNG channel samples the cGMP concentration in the adjacent cytoplasm once every few milliseconds. For practical purposes, the channels respond immediately to the decline in cGMP in the cytoplasm, and they do this with a very steep concentration dependence, because of the cooperativity noted above. Closing of channels reduces the flow of Na ⁺ ions into the cell, reducing the dark current and making the membrane potential more negative (i.e. hyperpolarized). Simultaneously, the flux of Ca ²⁺ into the cell is reduced as well by the action of the Na ⁺ /Ca ²⁺ exchanger discussed above.

Slowing of neurotransmitter release

Communication of rods with downstream bipolar cells is by the release of the neurotransmitter, glutamate. High levels of glutamate release by rods signal total darkness to bipolar cells, and reductions in the levels of glutamate release signal absorption of light. This change results from the hyperpolarization of the cell membrane, which is passively propagated along the plasma membrane from outer segment to synaptic terminal. Because rod cells are relatively short as compared to neurons with long axons, active propagation of the signal by action potentials is not needed to communicate potential changes at the outer segment to the synapse.

Rhodopsin phosphorylation, retinoid recycling and regeneration

Because of the catalytic activity of MII, as long as MII survives intact downstream signaling will continue, and cGMP will be subject to rapid degradation. There are multiple mechanisms for inactivating MII, but the fastest-acting and most important one appears to be phosphorylation by rhodopsin kinase, RK or RK1 ( Fig. 18.12 ). Like the G-protein, RK preferentially recognizes the activated form of rhodopsin, MII. RK uses ATP to attach phosphate groups to serine residues in the carboxyl-terminal tail of rhodopsin. Up to eight phosphates may be added, and if any of the key residues are missing, MII inactivation kinetics are slowed. ^, Phosphorylation of MII reduces its affinity for the G-protein, and slows the activation of the G-protein. Rhodopsin kinase is a member of a family of kinases specific for activated forms of G-protein-coupled receptors, which includes a kinase found only in cones (but not found in all species with cones), RK7. ^, RK-1 associates with the disk membranes via a covalently attached farnesyl group, ^, and RK7 via a geranylgeranyl group. The greater hydrophobicity of the 20 carbon geranylgeranyl group as compared to 15 carbon farnesyl moiety may enhance the efficiency of interactions between RK7 and photoexcited cone pigments.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Neural Control of Eye Movements
2. Metabolic Interactions between Neurons and Glial Cells
3. Electroretinogram of Human, Monkey and Mouse
4. The Effects of Visual Deprivation After Infancy

Stay updated, free articles. Join our Telegram channel

Join