"I Have Seen the Light!"
Vision and Light-Induced Molecular Changes

Spectroscopy and Quantum Chemistry Experiment

Authors: Rachel Casiday and Regina Frey
Department of Chemistry, Washington University
St. Louis, MO 63130


Key Concepts


Overview of the Vision Process

For most of us, vision is such an everyday occurrence that we seldom think to wonder how we are able to see the objects that surround us. Yet the vision process is a fascinating example of how light (such as the light reflected off of the objects that we see) can produce molecular changes with important consequences (i.e., our ability to perceive an image). The eyes receive the light and contain the molecules that undergo a chemical change upon absorbing light, but it is the brain that actually makes sense of the visual information to create an image. Hence, the visual process requires the intricate coordination of the eyes and the brain. How do these organs work together in order to allow us to see the light-reflecting objects around us as a visual image?

From the Light Source to the Brain: Mapping an Image

The eyes behave similarly, in some respects, to a camera. Light enters the pupil, is focused by the lens, and strikes a light-sensitive detector (called the retina) located along the inner surface of the back of the eye (Figure 1).

Figure 1

This is a schematic drawing of the human eye. Light enters the front of the eye through the pupil and is focused by the lens onto the retina. Rod cells on the retina respond to the light and send a message through the optic nerve fiber to the brain.

The light is mapped as an image along the surface of the retina by activating a series of light-sensitive cells known as rods and cones. These photoreceptor cells convert the light into electrical impulses which are transmitted to the brain via nerve fibers. For an image to be recognized, many photoreceptor cells will be activated and the visual information will be transported to the brain via numerous nerve fibers. The brain then determines, according to which nerve fibers carried the electrical impulse, which photoreceptors were activated by the light, and then creates a picture (Figure 2).

Figure 2

This figure shows how the brain uses mapping to make sense of visual information from the eye. The green numbers in the figure correspond to the following steps:

  1. Rays of light (blue) reflected off of an image are focused through the lens onto the back of the eye, forming an upside-down image on the retina.
  2. On the retina, those photocells that are hit by light from the image are activated. These photocells are shown in white in this figure. Photocells that do not receive any reflected light are not activated, and are shown in this figure. Thus, we can think of the image as a pixellate map of activated and nonactivated photocells on the retina.
  3. A nerve (gold) from each photocell connects to a particular location in the visual cortex of the brain. The photocells that are activated (white) send a nerve impulse to the brain, while the photocells that are not activated (black) do not send any impulse to the brain. (Only a small sample of the nerves are shown in this figure.)
  4. The brain, when it receives a collection of nerve signals from the eye, interprets where each signal comes from, and reconstructs the pixellate map.
  5. The brain then interprets the pixellate map as an image.

Photoreceptor Cells and Nerve-Impulse Generation

As explained above, the vision process is initiated when photoreceptor cells are activated by light from an image. Hence, our discussion of the vision process shall focus on the photoreceptor cells, and how these cells are activated to generate a nerve impulse to the brain.

The retina is lined with many millions of photoreceptor cells that consist of two types: 7 million cones provide color information and sharpness of images, and 120 million rods (Figure 3) are extremely sensitive detectors of white light to provide night vision. (The names of these cells come from their respective shapes.) The outer segments (tops) of the rods and cones contain a region filled with membrane-bound discs, which contain proteins bound to the chromophore 11-cis-retinal. (A chromophore is a molecule that can absorb light at a specific wavelength, and thus typically displays a characteristic color.) When visible light hits the chromophore, the chromophore undergoes an isomerization, or change in molecular arrangement, to all-trans-retinal (see below for a fuller description of this isomerization). The new form of retinal does not fit as well into the protein, and so a series of conformational changes in the protein begins. As the protein changes its conformation, it initiates a cascade of biochemical reactions that result in the closing of Na+ channels in the cell membrane, as outlined in Figure 4. Prior to this event, Na+ ions flow freely into the cell to compensate for the lower potential (more negative charge) which exists inside the cell. When the Na+ channels are closed, however, a large potential difference builds up across the plasma membrane (inside the cell becomes more negative and outside the cell becomes more positive). This potential difference is passed along to an adjoining nerve cell as an electrical impulse at the synaptic terminal, the place where these two cells meet. The nerve cell carries this impulse to the brain, where the visual information is interpreted.

Figure 3

This is a schematic diagram of a rod cell. The stacked disks contain rhodopsin, the complex of opsin protein and 11-cis-retinal. At the synaptic body, the potential difference generated as the ultimate result of the retinal isomerization is passed along to a connecting nerve cell, creating an electrical impulse that will be transmitted to the brain and interpreted as visual information.

Figure 4

This is a flowchart outlining the major steps in the vision signal transduction cascade which occurs between the isomerization of retinal (which leads to the formation of metarhodopsin II, the first reactant in the process outlined in this figure) and the interpretation of a visual image by the brain. The steps in this cascade are discussed in the section entitled "Signal Transduction Cascade to Generate a Nerve Impulse", below.

The Vision Process for Monochrome Vision (Rod Cells)

The sequence of events to generate a signal to the brain for monochrome vision (which occurs in the rod cells) and for color vision (which occurs in the cone cells) is essentially the same, although monochrome vision is somewhat simpler. Hence, we shall first describe how a monochromatic visual nerve impulse is generated, and then show how color vision differs. The process of generating a monochromatic visual signal can be broken down into three important steps: the isomerization of retinal, the protein conformational changes following retinal isomerization, and the signal transduction cascade to generate a nerve impulse.

Isomerization of Retinal

The first step in the monochrome vision process, after light hits the rod cell, is for the chromophore 11-cis-retinal to isomerize to all-trans-retinal. This event is best understood in terms of molecular orbitals, orbital energy, and electron excitation. You may find it helpful to review these important concepts in the introduction to the Experiment in your lab manual, or in your chemistry textbook.

In the Experiment, you learned that when an atom or molecule absorbs a photon, its electrons can move to higher-energy orbitals, and the atom or molecule makes a transition to a higher-energy state. In retinal, absorption of a photon promotes a p electron to a higher-energy orbital (a p-p* excitation). This excitation "breaks" the p component of the double bond, thus allowing free rotation about the bond between carbon atom 11 and carbon atom 12 (see Figure 5). Thus, when 11-cis-retinal absorbs a photon in the visible range of the spectrum, free rotation about the bond between carbon atom 11 and carbon atom 12 can occur and the all-trans-retinal can form. This isomerization occurs in a few picoseconds (10-12 s) or less. Energy from light is crucial for this isomerization process: absorption of a photon leads to isomerization about half the time; in contrast, spontaneous isomerization in the dark occurs only once in 1000 years! The molecule resulting from the isomerization is called all-trans-retinal. In the cis configuration, both of the attached hydrogens are on the same side of the double bond; in the trans configuration, the hydrogens are on opposite sides of the double bond. As you can see by the CPK representation in Figure 5, the cis-trans isomerization causes the conjugated carbon chain (alternating double and single bonds) to become straightened, and increases the distance between the -CH3 group attached to carbon 5 and the oxygen at the end of the chain.

11-cis-retinal

all-trans-retinal

Figure 5

Upon absorption of a photon in the visible range, 11-cis-retinal can isomerize to all-trans-retinal. In the 11-cis isomer, the hydrogens (red in the 2-D ChemDraw representation) are on the same side of the double bond (red in the 2-D ChemDraw representation) between carbon atom 11 and carbon atom 12. In the all-trans isomer, the hydrogens are on opposite sides of the double bond. In fact, all of the double bonds are in the trans-configuration in this isomer: the hydrogens, or hydrogen and -CH3, are always on opposite sides of the double bonds (hence, the name "all-trans-retinal"). Note how the size and shape of the molecule change as a result of this isomerization.

Note: On the left is shown the two-dimensional ChemDraw representation of the isomerization reaction. On the right are CPK models of the two isomers. Carbon atoms are gray, hydrogens are light blue, and oxygens are red in these CPK models. The CPK images were rendered using SwissPDB Viewer and POV-Ray (see References).

Note: To view the molecule interactively, please use Jmol, and click on the button under the isomer that you wish to view.


Questions on the Isomerization of Retinal


Protein Conformational Changes Following Retinal Isomerization

As we shall see below, the isomerization of retinal has an important effect on special proteins in the rod cell: the isomerization event actually causes the proteins to change their shape. This shape change ultimately leads to the generation of a nerve impulse. Hence, the next step in understanding the vision process for monochrome vision is to describe these proteins, and how they change their shape after retinal isomerizes.

In rod cells, the protein which binds the chromophore retinal is opsin, and the bound complex of 11-cis-retinal plus opsin is known as rhodopsin, or visual purple. Alone, 11-cis-retinal has a maximum absorbance in the ultraviolet part of the spectrum, but the maximum absorbance for rhodopsin is 500 nm (in the visible green part of the spectrum). Recall from the inorganic-synthesis experiment that the observed color of a substance is actually the complementary color to the color that is absorbed. Thus, the name "visual purple" describes the complementary color for rhodopsin. (Rhodopsin also absorbs in the ultraviolet region of the spectrum. However, the lens of the eye absorbs ultraviolet light, preventing it from reaching rhodopsin in the retina. This is why we cannot see ultraviolet light.) Opsin consists of 348 amino acids, covalently linked together to form a single chain. This chain has seven hydrophobic, or water-repelling, alpha-helical regions that pass through the lipid membrane of the pigment-containing discs. (You will learn more about membranes and membrane-spanning proteins in the Chem 152 tutorial, "Maintaining the Body's Chemistry: Dialysis in the Kidneys".) This region consists primarily of nonpolar amino acids, which do not attract the polar water molecule. (The alpha-helical structural motif is described in detail in the tutorials "Hemoglobin and the Heme Group: Metal Complexes in the Blood for Oxygen Transport" and "Iron Use and Storage in the Body: Ferritin and Molecular Representations".) The chromophore is situated among these alpha helices in the hydrophobic region. It is covalently linked to Lysine 296, one of the amino acids in the opsin peptide chain (Figure 6).

Figure 6

The upper panel is a two-dimensional representation of the reaction which links 11-cis-retinal to opsin. The lower panel is a three-dimensional close-up of Lysine 296 covalently attached to 11-cis-retinal.

Note: The coordinates for the three-dimensional representation were obtained using molecular modeling. The the image was rendered using SwissPDB Viewer and POV-Ray (see References).

When the chromophore absorbs a photon it isomerizes to the all-trans configuration without (at first) any accompanying change in the structure of the protein (Figures 7 and 8). Rhodopsin containing the all-trans isomer of retinal is known as bathorhodopsin. However, the trans isomer does not fit well into the protein, due to its rigid, elongated shape. While it is contained in the protein, the all-trans chromophore adopts a twisted conformation, which is energetically unfavorable. Therefore, a series of changes occurs to expel the chromophore from the protein.

Figure 7

These are schematic diagrams of rhodopsin (11-cis-retinal bound to opsin) and bathorhodopsin (all-trans-retinal bound to opsin) in the membrane of a pigment-containing disc in the rod cell. 

Figure 8

These are views from above of the seven helices of opsin bound to retinal, before and after the cis-trans isomerization. The all-trans-retinal isomer does not fit well into the protein, and so it will cause a series of conformational changes by which the chromophore is removed from the protein.

Note: To view the opsin protein in its two conformations interactively, please use Jmol, and click on the "View this Molecule Interactively" button under the conformation that you wish to view.

The coordinates for the transmembrane helices and chromophore shown in Figures 7 and 8 were obtained from the THEORETICAL MODEL by Pogozheva et al. The trans-membrane helices are depicted using the ribbon representation to emphasize the helical structure while the chromophore is depicted using the CPK, or spacefilled, representation to show the approximate volume occupied by retinal. Helical portions of the images were rendered using SwissPDB Viewer and POV-Ray (see References). The membrane and non-helical protein segments in Figure 7 are cartoon representations.  Click to view the EXPERIMENTALLY DETERMINED X-ray crystal structure of rhodopsin, which was recently published in May of 2000 (see References). 

Although the initial isomerization occurs without any change in the shape of the opsin protein, the twisted conformation of all-trans-retinal in bathorhodopsin is too unstable to remain in this configuration for long. Within nanoseconds (10-9 s), the shape of the protein begins to change. Ultimately, the all-trans-retinal molecule is expelled from the protein, yielding free opsin plus free all-trans-retinal. A series of intermediate complexes have been isolated at low temperatures, each absorbing maximally at a different wavelength. The names and characteristic lmax values for these intermediates are shown in Table 1.

Pigment Name

lmax

Rhodopsin

Bathorhodopsin

Lumirhodopsin

Metarhodopsin I

Metarhodopsin II

trans-Retinal (free)

498 nm

543 nm

497 nm

487 nm

380 nm

370 nm

Table 1

As rhodopsin changes its conformation following the cis-trans isomerization of retinal, the intermediates have distinct names and absorbance characteristics.

Source: Yoshizawa and Kuwata on bathorhodopsin, Cowan and Drisko on other intermediates.

For vision to occur, the important intermediate is metarhodopsin II. This intermediate activates the enzyme transducin, which starts the impressive signal transduction cascade resulting in the production of a nerve impulse to the brain (see next section). The all-trans-retinal is important for the multiple-step regeneration of 11-cis-retinal. The regenerated 11-cis-retinal is re-incorporated into rhodopsin.


Questions on Protein Conformational Changes Following Retinal Isomerization


Signal Transduction Cascade to Generate a Nerve Impulse

After the metarhodopsin II is formed, there are approximately four more steps in the vision process: activation of the enzymes transducin and phosphodiesterase, hydrolysis of cyclic GMP, closing of Na+ channels, and propagation of an electrical impulse to the brain (see the flowchart in Figure 4).

Recall, for a signal to be sent via a nerve fiber, the Na+ channels must be closed so that a large charge difference across the rod's outer membrane builds up. (Recall the plasma membrane shown in Figure 3.) Once a large charge difference occurs, the membrane is said to be hyperpolarized. Then, charge travels as an electrical impulse down the rod cell to the synaptic terminal, where it is transferred to an adjoining nerve cell. How does this charge buildup occur?

The starting point for this process is the production of metarhodopsin II, as described above. This initiates the following cascade of events: First, metarhodopsin II complexes with the enzyme transducin and activates it. Transducin in turn activates another enzyme, phosphodiesterase. Phosphodiesterase catalyzes the hydrolysis of cyclic GMP, as shown in Figure 9, below.

Figure 9

Phosphodiesterase catalyzes the hydrolysis of cyclic GMP, adding a water molecule (red) to GMP and breaking the bond between the phosphate group (blue) and carbon.

Note: The "cyclic" part of cyclic GMP (on the reactant side of this reaction) is composed of the phosphate group (blue) and the three carbon atoms between the oxygens of the phosphate. The GMP on the product side of this reaction is not considered to be cyclic.

Cyclic GMP is required to open Na+ channels in the plasma membrane. In the dark, cyclic GMP is abundant and these channels stay open. Sodium cations enter freely into the rod cell, because the cell typically has a lower potential (is more negative) than the external environment, thus attracting the positively-charged ions. However, when cyclic GMP is hydrolyzed (gains an H2O and breaks a bond) by the now-activated phosphodiesterase, it is no longer available to keep the Na+ channels open. Sodium cations can no longer enter the cell freely, and so the cell's potential suddenly becomes even lower relative to the external environment. A large charge difference across the membrane is built up; this is known as hyperpolarization. The large potential difference travels as an electrical impulse down the rod cell to the synaptic terminal, and is then transferred to an adjoining nerve cell. The nerve cell carries this impulse all the way to the brain. The brain then determines where the nerve impulse originated, and interprets the image, as shown in Figure 2.


Question on the Signal Transduction Cascade to Generate a Nerve Impulse

  1. If the scientist injects a chemical that blocks the activity of transducin into the rod cells, what effect (if any) would this injection have on vision? Briefly, explain your answer.
  2. If the scientist injects activated phosphodiesterase into the rod cells, what effect (if any) would this injection have on vision? Briefly, explain your answer.

Color Vision (Cone Cells)

Now that we have studied the vision process for monochrome vision, we can turn our attention to color vision. Recall that the nerve signals for color vision are generated in the cone cells. Color vision in the cone cells operates by essentially the same process as the monochrome vision in the rod cells. However, whereas the eye only has one type of rod cell, the eye has three different types of cone cells. The differences between the three types of cone cells, as we shall see below, allow us to distinguish colors.

Our color vision is trichromatic, i.e., we perceive color through three fundamental receptors: red-absorbing, green-absorbing, and blue-absorbing cone cells. Each color in the visible spectrum can be made by a mixture of these three primary colors recognized by the three types of cone cells. Each type of cone cell contains a different protein bound to 11-cis-retinal and has its own characteristic absorption spectrum, corresponding to the particular pigment protein that it contains. Each absorbs maximally at a characteristic wavelength, but the absorbance peaks are rather broad and extend over a range of wavelengths. Therefore, some wavelengths are absorbed (to varying degrees) by more than one type of cone. Orange light, for instance, is absorbed by both the green- and red-absorbing pigments, but the latter pigments absorb the orange light more efficiently. When the brain receives the combination of signals from the green- and red-absorbing cones, it interprets this as orange light. The combination would be different for yellow light (more green relative to the red absorbance). The three pigment proteins for color vision are similar to rhodopsin, and contain much of the same amino-acid sequence as rhodopsin. Only a few amino acids located near the retinal binding site are varied in these proteins. The absorption spectra of the red- and green-absorbing proteins are tuned by the presence of amino acids containing a hydroxyl (-OH) group near the retinal binding site. Recall that the alpha helices where retinal binds are hydrophobic, composed of nonpolar amino acids. The hydroxyl group is a polar group (consider the relative electronegativities of oxygen and hydrogen), and therefore more attractive to water. It has been shown that replacement of a nonpolar amino acid with a polar amino acid at certain key positions shifts lmax about 10 nm to longer wavelengths (lower energy). The red and green proteins have more of their amino acid sequence in common than either does with the blue protein. Therefore, their absorption spectra are similar, whereas the absorption spectrum of the blue-absorbing pigment is more distinct.


Questions on Color Vision

  1. Label each peak to show which type of receptor (i.e., red-absorbing, green-absorbing, or blue-absorbing) it represents.
  2. The wavelength for indigo light is 480 nm. Which cone(s) in the goldfish eye would produce a nerve signal for indigo?
  3. Assume that the absorption spectra for human cone cells are similar to those of goldfish cone cells. If ultraviolet light were not filtered out by the lens of the eye, would we perceive ultraviolet light as color? Which cone(s) would absorb ultraviolet light? Briefly, explain your answer.

Comparison of Rods and Cones: Sharpness and Sensitivity

Now that we have discussed how light excites vision cells to generate a nerve impulse, and how different cone cells allow us to differentiate color, the question remains: why are rod cells more sensitive to small amounts of light, and why do cone cells provide sharper images? The answer lies in the brain's ability to map images according to the location of the photoreceptor cells that send nerve impulses to the brain (recall Figure 2). The brain has no direct contact with the photoreceptor cells, but receives information through an intermediary optical nerve. Thus, our ability to view images depends on the brain determining the location of the photoreceptor cell that passes an impulse to any given nerve fiber.

Each cone cell connects to a different nerve fiber, so the brain is able to precisely determine the location of the visual stimulus. Thus, cone cells provide very sharp visual images. Rod cells, however, may share a nerve fiber with as many as 10,000 other rod cells. When the brain receives an impulse from such a nerve fiber, it has no way of determining from which of these 10,000 cells the impulse originated. The brain cannot therefore distinguish the precise location of the rod cell sending an electrical impulse, so the image is not as sharp as that provided by cones. On the other hand, because the signals from rods are pooled and not spread among as many nerve fibers, rods provide greater sensitivity to a very small light response, such as occurs in very dim light.

Summary

Vision is a remarkable process by which we are able to interpret an image from light the eyes receive from the objects around us. Although this process depends on the interplay of many different factors (including the optics of the eye, the isomerization of retinal, nerve impulses, and the brain's ability to reconstruct the image), vision is fundamentally based on the change in the molecular orbitals of retinal that occurs when the molecule absorbs energy in the form of light reflected off of the objects that we see. Recall that when visible light hits the chromophore (retinal), a p electron is promoted to a higher-energy orbital, allowing free rotation about the bond between carbon atom 11 and carbon atom 12 of the retinal molecule. About half the time, this rotation leads to the isomerization of retinal when the p electron returns to the lower-energy orbital. When retinal isomerizes, a conformational change in the protein opsin occurs. This conformational change initiates a cascade of biochemical reactions that result in the closing of Na+ channels in the cell membrane. When the Na+ channels are closed, a large potential difference builds up across the plasma membrane, and the potential difference is passed along to an adjoining nerve cell as an electrical impulse. The nerve cell carries this impulse to the brain, where the visual information is interpreted.


Jmol Files

To view the molecules interactively, please use Jmol. To download the pdb files for viewing and rotating the molecules shown above, please click on the appropriate name below.


Additional Links:


References:

Cowan, D.O. and R.L. Drisko. Elements of Organic Photochemistry, New York: Plenum Press, 1976, p. 561-565.

Guex, N. and Peitsch, M.C. Electrophoresis, 1997, 18, 2714-2723. (SwissPDB Viewer) URL: http://spdbv.vital-it.ch/.

Nathans, J. "The Genes for Color Vision," (1987) Scientific American, 255 (7), 42-49.

Okada, T., I. Le Trong, B. A. Fox, C. A. Behnke, R. E. Stenkamp, and K. Palczewski. "X-Ray Diffraction Analysis of Three-Dimensional Crystals of Bovine Rhodopsin Obtained from Mixed
Micelles," Journal of Structural Biology, Vol. 130, No. 1, May 2000, pp. 73-80.

Persistence of Vision Ray Tracer (POV-Ray). URL: http://www.povray.org.

Pogozheva, I.D. et al. "The transmembrane 7-alpha-bundle of rhodopsin: Distance geometry calculations with hydrogen bonding constraints," (1997) Biophys. J, 72, 1963-85. Rhodopsin PDB coordinates, Brookhaven Protein Data Bank.

Sappan, P. Chemistry and Light. Royal Society of Chemistry, Cambridge, 1994, p. 171-175.

Stryer, L. "The Molecules of Visual Excitation," (1987) Scientific American, 257 (7), 42-50.

Stryer, L. Biochemistry, 4th ed., W.H. Freeman and Co., New York, 1995, p. 332-339.

Wald, George. "The Molecular Basis of Visual Excitation," (1968) Nature, 219, 800-807.

Yoshizawa, T. and O. Kuwata. Organic Photochemistry and Photobiology, W. Horspool and P. Song, eds. Ch. 26, "Vision: Photochemistry," p. 1493-1499. CRC Press, New York, 1995.


Acknowledgements:

The authors thank Greg Noelken for creating the chime script files. They also wish to thank Dewey Holten, Carolyn Herman, Michelle Gilbertson, and Jody Proctor for many helpful suggestions in the writing of this tutorial.

The development of this tutorial was supported by a grant from the Howard Hughes Medical Institute, through the Undergraduate Biological Sciences Education program, Grant HHMI# 71199-502008 to Washington University.

Copyright 1998, Washington University, All Rights Reserved.

Revised November, 2000.