Molecular Cell Physiology - MCDB 3280
3280 Home

Unit 14b - Rhodopsin

  • Retinal (the aldehyde form) is a derivative of vitamin A (the -OH form), which itself is a derivative of beta-carotene, the pigment that gives carrots their characteristic color and also is found in pumpkins, spinach, squash, watermelon, asparagus, broccoli, cantaloupe, and liver.
    • Vitamin A is an essential component for vision and it promotes bone growth, tooth development, and helps maintain healthy skin, hair, and mucous membranes. Deficiencies of vitamin A result in a number of maladies including night blindness, dry skin, poor bone growth, weak tooth enamel, and weight loss.
  • The covalent linkage is to the lysine at position 296.

  • Opsin is a 348 aa, 40 kD protein
  • Mutation of either proline at position 23 or glutamine at position 28 causes the disease retinitis pigmentosa (RP)
    • RP is an inherited disease in which photoreceptors slowly degenerate, starting with rods, leading to "legal" blindness by about age 40. Numerous different defects in rhodopsin are causes of RP.
    • Numerous other genes involved in photoreceptor function also cause RP.
      • The most common group of genes mutated in RP are those encoding proteins of the visual (g-protein) cascade of the photoreceptor cell outer segment.
      • The second common group of genes mutated in RP are those encoding proteins of the visual cycle, the series of biochemical steps that provide and recycle retinal.
    • Inappropriate protein folding may be one cause of pathogenesis. The mutations in aa-23 or aa-28 may be involved in such misfolding.
      • the mutant protein forms aggregates that lead to the dispersal of the Golgi and disrupt the intermediate filament network.

Here are pictures of the rhodopsin crystals of the type used to determine the crystal structure seen below. The top set are dark adapted rhodopsin that was isolated in dim red light that does not readily isomerize rhodopsin. the bottom figure is after one minute of bleaching. You can see why the original name of rhodopsin was 'visual purple'.

X-ray structure of PDB file 1F88

The -COOH terminal is on the intracellular face which corresponds to the cytoplasmic side of other GPCRs.

This backbone tracing emphasizes the fact that the helices are tilted. (Red line is perpendicular to membrane plane.) Such tilting appears to be a common feature of many membrane proteins. One possible explanation is that it provides for flexing and shape change in response to ligands.

Also related to the helices is that they are not perfect alpha-helixes. this probably provides flexibility for the conformational change after photon absorption. For example there is a proline at position 53 in helix 1 that causes a kink that disrupts normal alpha-helical conformation.

The next figure shows the position of the membrane borders in blue:

The loops at the intradiscal face (the extracellular face in other GPCRs) essentially form a plug at that end of the molecule (two pictures up). The plug abuts the 11-cis-retinal pocket. It prevents the all-trans form from sticking out, instead assisting the significant change in helix packing that occurs after photoisomerization.

It is expected that this 'extracellular face is not plugged in other GPCRs because it is the entrance for the ligand that will bind with the GPCR.

  • This can be seen in this model of the beta1-adrenergic receptor that was prepared as a simulation based on the known sequence of the receptor and of the solved structure of rhodopsin:

Intra-helical hydrogen bonds at the level shown by the horizontal line through the above structure help maintain the structural retlationship among the helices. These are seen in end-on view in the next picture,

Another highly conserved region in GPCRs is near the cytoplasmic face and also hydrogen bonds the helices

Chromophore Pocket

x-ray crystallography solution of 11-cis-retinal shape seen in cross section

Above: cut-away spacefill model.
Below: key side groups lining the pocket

The next figure emphasizes the interactions that stabilize the Schiff-base linkage to lysine 268:

The next figure shows a more intricate hydrogen bonded scheme around the pocket when two water molecules that lie inside the center of the helices are taken into account:

The cytoplasmic (inter-disc) portion of the molecule

  • The C-IV loop is helical with hydrophobic side chains that hold it parallel to the membrane. It is a key point of contact with the g–protein.
    • The opposing position of the hydrophilic and hydrophobic side chains can be seen in this end-on representation of the alpha-helix of the loop

  • The structure solved so far is the 11-cis, inactive form. the change to the all-trans form causes a steric rearrangement of the transmembrane helices that is reflected in a change at the cytoplasmic face. Prior to the change the inactive form does not bind the g-protrotein (transducin) and after the change it does.
  • The exact nature of the activational change awaits a crystal-structure detrmination of the all-trans form of rhodopsin. However, various experiments as well as comparison to the solved structure of bacteriorhodopsin suggest the general nature of the changes.
  • The next picture shows the changes expected as the isomerization occurs, based on computer simulations. (a) is the 11-cis forms and b-d are at increasing times up to 6 ns as the chromophore is changing shape.

The next figure shows the activated form of the all-trans molecule (known as metarhodopsin-II) based on biochemical experiments and calculations, compared with 11-cis rhodopsin.

  • Note the change in relative position of the transmembrane helices, as well as the change in the conformation of the intra-cytoplasmic loops. A potential "binding groove" opens up.

  • In bacteriorhodopsin the changes are shown in the next pictures. The lower figure shows the change in the helox represented in red (preisomerization superimposed as grey), as well as twists in the side groups of nearby sidechains.

Binding to the G-protein

Biochemical experiments suggest that in the 11-cis configuration, cytoplasmic loop IV is in the helical configuration discussed above (shown again below) bound to membrane phosphotidylserine, and in the all-trans configuration it part of it changes to an open loop configuration because of loss of the hydrophobic binding site due to shape changes of the rhodopsin.

  • Mutation studies suggest that the binding of loop 3 to region A of the transducin alpha-subunit is one of the mechanisms of specific G-protein activation by bovine rhodopsin.
    • Region A is a 6-amino acid alpha-helix sequence (yellow, below) at a position adjacent to the C-terminal 5 amino acids of the G-alpha subunit.

  • Other studies indicate that loops II and III, and nonloop cytoplasmic regions are involved in the specific interaction with the alpha-subunit of the G-protein.

Wavelength Sensitivity

What is the basis of the wavelength specificity of rhodopsin and the other photoreceptor visual pigments?

  • Wavelength specificity means that there is a best wavelength and overall wavelength envelope that characterizes the absorption spectrum of rhodopsin and cone opsins.

  • The absorption spectrum of rhodopsin with a peak at 498 nm (or of the cone opsins) cannot simply be due to the absorption spectrum of 11-cis-retinal, which aborbs in the ultraviolet with a peak at 380 nm.
  • So, what shifts the rhodopsin into the visable with a peak absorption at 498 nm?

Basis of Wavelength Tuning in Rods (Rhodopsin)

A side-trip into organic chemistry and quantum mechanics (to prove that all that Chemistry you took was relevant after all).

(Much of this section draws on the pictures and explanations at these web sites:
Organic Chemistry Online, chapter 1
Particle In A Box from the Chemistry Hypermedia Project
One-Dimensional Particle in a Box Applications from Univ. of Guelph
The Particle In A Box applet from UC-Irvine)

1. Start by recalling that the electrons of an atom can only occupy certain energy states in the vicinity around the nucleus. That is, the allowable energy states are quantized due to the wavelike nature of an electron.

The UC-Irvine site noted above provides a nice analogy of why only some energy levels are possible, which is quoted here almost verbatim. It starts with the realization that the wavelike aspect of an electron confined to a circular path must be a standing wave. that is, the wave must come back to the same point at the end of the path, which is the same point as the start of the path. Then, this is likened to waves in a bathtub, as follows:

  • "if you swish your hand around in the bathtub, you'll note that the only waves that persist for a long time are those that "fit" inside the bathtub, in the sense that a whole number of waves, measuring from crest to crest, fit into the bathtub.
  • "These waves are called 'standing' waves because, well, they just stand around instead of travelling forward like waves at the beach. In other words, they have their crests and troughs always in the same places, and the areas midway between the crests and troughs (the 'nodes') where the water level is undisturbed also always in the same places. Here's the largest possible bathtub wave, which has a crest at each end:

  • "Here's another, which has a crest at each end and in the middle:

  • "The possible energies of standing waves in the bathtub are limited to certain, specific (quantum!) values by the requirement that each wave fit exactly into the bathtub.
  • "Now the energies of the bathtub waves, like the wavefunctions of the particle in the box, increase with the number of wave crests that fit into the tub. If you swish the water very gently and slowly, for example, then the only wave you'll see will be the ``sloshing'' wave that has a crest exactly at each end of the tub and a node in the middle. As you increase your agitation -- whee! -- you'll see waves that have more and more crests in the middle, at least until you slop all the water out onto the floor. . . "

The last bullet can be restated tha to get from one allowed quantum level to a higher one, an input of energy is needed. AND, the amount of energy must be equal to the difference in the levels, not a bit less or a bit more, or it is not absorbed.

2. Next, recall that a hydrogen atom can make one covalent bond with another atom.

  • The single electron of each hydrogen atom occupies the lowest energy state, the s–orbital, and a bond between two hydrogen atoms is thus an s-bond.

  • Because two atomic orbitals were combined, there must be two molecular orbitals, the lower energy bonding orbital and a higher energy anti-bonding orbital.

  • A carbon atom has four electrons and thus four orbitals occupied, the s–orbital, and three perpendicular p–orbitals.

  • When a C=C double-bond is made, one of the bonds is between the s–orbitals and the other is between two p–orbitals, such that the electrons in the p–orbital overlap to form a pi–bond cloud above and below the plane of the s–bond.

  • Further, a double bond, or a chain of conjugated double bonds must be planar. There cannot be rotation around the s-bond because then p-bond overlap would not be possible.
    • This is why the retinal molecule is planar.

  • Also, analogous to the H-H case discussed above, there must be two more orbitals of higher energy in the C=C bond, the pi-anti-bonding orbitals. An electron is elevated to the first anti-bonding orbital, signified by pi*, if exactly the correct amount of energy is absorbed. And when that happens, rotation around the s-bond can then occur.
    • This is how the rotation around the 11-12 cis-bond in retinal occurs when the energy of a photon is absorbed, allowing movement to the 11-12 trans configuration.
  • In a conjugated bond molecule such as retinal the representation below is purely diagrammatic.

  • Just as in the familiar case of benzine, the single-double bond character is shared all along the chain of single and double bonds into one overall resonance structure.

3, It is the overall length of the resonance bond that governs the energy states between the combined orbital that it occupies and the next highest, unfilled anti-bonding orbital that can be occupied when a photon is absorbed. The calculation of the distance between these two orbitals is made using the quantum mechanical Schroedinger Equation.

  • Bypassing lots of theory and math, the equation for this energy difference, delta E, looks something like this, where L is the Length of the conjugated resonance bond:

delta E = constant/L*2

  • This energy corresponds to a wavelength. the relationship is a inverse one, such that short wavelengths, like x-rays, have high energy and long wavelengths have lower energy, such as infrared. So the wavelength that correspoonds to delta E, call it Lambda is:

Wavelength of absorbed photon = Lambda = L*2/constant

  • Translated into words, this says that the "longer" the resonance bond, the longer the wavelength of the photon that is absorbed.
    • The reason the absorption spectrum is a spread out curve, rather than a single narrow peak is because there are vibrational and translational states "around" the best state that give a probability of wavelengths distributed around the peak at which absorption can occur.

4. Finally we can answer the question posed above, what shifts rhodopsin into the visable with a peak absorption at 498 nm, when 11-cis retinal absorbs at around 380 nm?

  • The 'Length' referred to above is actually a 3-dimensional cloud that was simplified into a one-dimensional length for purposes of illustration. So, really it is the overall space occupied by the resonance bond 'cloud' that governs the wavelength of the absorbed cloud.
    • Still, the larger the volume of the cloud, the longer the wavelength that is absorbed.
  • Some of the amino acid sidegroups that make up the pocket in rhodopsin where the retinal lies are charged. The net result of this is to delocalize the 'resonance cloud' and make it larger. This shifts the wavelength of the photon that can be absorbed from a short wavelength of about 380 nm in the ultraviolet, to the longer wavelength of 498 nm in the visible!! In this picture some of the polar or charged side groups within 4.5 Angstroms of the retinal are highlighted in green.

Cones and Color Vision

Typically, a mammal with color vision has three different cones of long (L red), medium (M green) and short (S blue) wavelength sensitivity.

  • The different sensitivities are due to modest differences in the opsin sequence that change the side groups in the pocket. Key, charged amino acids that are important in this tuning are shown in red in the next figure.

  • The next figure illustrates important amino acid differences in a different representation.

  • The various opsins make up an extremely large group of GPCRs. Their evolutionary relationship has be proposed based on the sequence changes and looks something like this:

    • Different animals have different sensitivities that aparently evolved in response to their environment.
      • For example deepwater fish and shallow water fish of similar species have differing cone spectra as does their environment due to the absorption characteristics of water.
      • It has been proprosed that the peaks in the spectra of monkey cones and the colors of the fruit they eat have co-evolved to match. In the simplest example, here is what a red-green colorblind (missing either L or M pigment) monkey would see; the fruit becomes hard to pick out from the leaves.

      • The next figure looks at the hypothesis more critically. It plots the long versus middle wavelength cones on the axes and the measured reflectance of the food eaten by the monkey (typically orange-red). the contours are fruit refflectance (white is greatest) and the X is the point of maximum sensitivity of the two cone types in this monkey species. Note the match.

  • Some humans, especially women (due to Barr body selection) actually have four different cone types each with a different spectrum. The richness of their color experience is greater than normal trichromats.
  • For example, the four-pigment individuals distinguish broader (and a bit different) chromatic bands in the rainbow than do normals

References

The paper that reports the crystal structure, 1F88, of rhodopsin: Krzysztof, P, et al. (2000) Crystal structure of rhodopsin: A G protein-coupled receptor. Science. 289:739-745.

The commentary on that paper in the same issue of Science: Bourne HR, Meng EC. (2000) Rhodopsin sees the light. Science. 289:733.

Choi G, et al. (2002) Structural Studies of Metarhodopsin II, the Activated Form of the G-Protein Coupled Receptor, Rhodopsin Biochemistry. 41(23):7318-7324 (full-text online)

Jameson KA, et al. (2001) Richer color experience in observers with multiple photopigment opsin genes. Psychometric Bulletin and Review. 8(2):244-261.

Krisha GA, et al. (2002) Evidence That Helix 8 of Rhodopsin Acts as a Membrane-Dependent Conformational Switch. Biochemistry. 41(26):8298-8309.Crystal structure of rhodopsin: implications for vision and beyond

Tetsuji Okada T, Palczewski K. (2001) Crystal structure of rhodopsin: implications for vision and beyond. Current Opinion in Structural Biology. 11:420–426

Phelan JK, Bok D. (2000) A brief review of retinitis pigmentosa and the identified retinitis pigmentosa genes. Molecular Vision. 6:116-24. (full-text online)

Regan BC, et al. (2000) Fruits, foliage and the evolution of primate color vision. Phiolsophical Transactions of the Royal Society of London, series B. 356:229-283.

Saliba, RS, et al. (2002) The cellular fate of mutant rhodopsin: quality control, degradation and aggresome formation. Journal of Cell Science. 115(14):2907-18.

Stenkamp RE, et al. (2002) Crystal structure of rhodopsin: a template for cone visual pigments and other G protein-coupled receptors. Biochimica et Biophysica Acta. 1565:168– 182.

Terakita A, et al. (2002) Functional Interaction between Bovine Rhodopsin and G Protein Transducin. Journal of Biological Chemistry. 277(1):40-46.

Vaidehi N, et al. (2002) Prediction of structure and function of G protein-coupled receptors. PNAS. 99(20):12622-7. (full-text online)

Yokoyama S. (2000) Molecular evolution of vertebrate visual Pigments. Progress in Retinal and Eye Research. 19(4):385-419.