Ferritin Molecular-Graphics Tutorial

Iron in Biology: Study of the Iron Content in Ferritin, The Iron-storage Protein.

Authors: Regina F. Frey, Maureen J. Donlin, and James K. Bashkin (1995)

Modified by Rachel Casiday and Regina Frey (1998)

Department of Chemistry, Washington University
St. Louis, MO 63130

Ref: M.J. Donlin, R.F. Frey, C. Putnam, J.K. Proctor, J.K. Bashkin, J. Chem. Edu., 75, 437, (1998)

Key Concepts

NOTE: Click on text that is highlighted in blue for more information on that topic.


Iron, an essential element in living organisms, is commonly used in the Fe(II) oxidation state. But in our oxidizing atmosphere, Fe(III) is the more prevalent oxidation state. At the physiological pH of 7, the Fe(III) ion concentration in aqueous solution is minimal. However, most organisms maintain an intracellular concentration of Fe(III) several orders of magnitude higher than simple aqueous solutions permit. This discrepancy in concentration demonstrates the striking ability of biochemical systems to concentrate and store iron. Conversely, iron can be very toxic, so the ability to store and release iron in a controlled fashion is essential. Cells have solved this problem of iron storage by developing ferritins, a family of iron-storage proteins that sequester iron inside a protein coat as a hydrous ferric oxide-phosphate mineral similar in structure to the mineral ferrihydrite.

In this laboratory experiment, the chemical properties of ferritin, the major iron-storage protein in living organisms, are examined. First, the total amount of iron in a ferritin sample is determined and then the time course of iron released from the protein by the reduction of the iron-mineral core is determined. In this manner, the mechanism of the in vitro iron-release process is examined.

Graphical Molecular Representations

The three-dimensional structure of a molecule is a critical determinant of its chemical properties. Thus, it is especially important, when studying the chemistry of large molecules (e.g., ferritin and other proteins), to visualize the three-dimensional structure of the molecules, and understand the relationship between structural features and molecular functions. Of course, there are many difficulties involved in converting all of the important structural information about a molecule into an easily understandable two-dimensional representation. No two-dimensional representation can show a three-dimensional structure in its entirety. Hence, a variety of molecular representation formats have been developed; each of these representations is designed to show a particular aspect of a molecule's structure. Thus, to illustrate a specific point about a molecule's structure, the type of representation must be chosen carefully. To provide a comprehensive view of a molecule's structure, multiple representations are used.

Graphical computer modelling has greatly improved our ability to represent three-dimensional structures. One of the goals of graphical computer modelling is to create the computer-generated image such that the image seems three-dimensional. By replicating the effect of light on three-dimensional objects, computers can give the impression of depth to simulate the three-dimensional aspect. The ability of interactive molecular viewing (e.g., using the RASMOL program) has enhanced our understanding of molecular structure even more, especially in the biochemical area. By interactively rotating the molecules, a clear picture of the three-dimensional structure emerges. In addition, this increases our chemical intuition by looking at two-dimensional images and visualizing the three-dimensional structure in our brains.

This graphical tutorial uses different types of structural representations, such as 2D-Chem Draw, stick, CPK, and ribbon, to illustrate the molecules used in this experiment. PDB files are also available for viewing the molecules interactively. As you view the tutorial, you should especially think about:

  1. The differences between the 3-fold and 4-fold channels found in ferritin;
  2. The size and shapes of the different Fe(II) compounds and how they relate to the channel sizes;
  3. The structure of the iron-mineral core and how the mineral core binds to the walls of the protein;
  4. The relationship between different types of structural representations such as two-dimensional sketch, stick, CPK, and ribbon.

Structural Overview of Ferritin

Ferritin is a spherical shell that consists of 24 subunits (or peptide chains) folded into ellipsoids, as shown in Figure 1. Each subunit is an individual molecule that joins to its neighboring subunits through noncovalent interactions; the subunits have a combined molecular weight of 474,000. These subunits pack to form a hollow sphere approximately 80 Angstroms in diameter with walls that are approximately 10 Angstroms thick. Among the important structural features of ferritin is the presence of two types of channels that occur in the protein wall at the intersection of the subunits.

Figure 1:

This is a molecular model of Ferritin with the subunits displayed in the CPK representation. CPK pictures represent the atoms as spheres, where the radius of the sphere is equal to the van der Waals radius of the atom. Hence, CPK representations are a good way to show the approximate volume occupied by a molecule.

All of the 24 subunits are identical, but they have been color coded to help illustrate the structure. Light blue subunits are closest to you, magenta subunits are farthest away, and dark blue subunits are inbetween. Coordinates for the model were determined from X-ray crystallographic data. The four (4) subunits colored in light blue form the walls of a 4-fold channel. The 3-fold channels occur at the intersection of the light blue, dark blue, and magenta-colored subunits. The location of 3-fold channels are indicated on the figure, but the channels themselves are obscured from this viewing angle.

View Interactively Note: To view this molecule interactively, please use Jmol, and click on the button to the left.

Questions on the Structure of Ferritin

Note: Some of the following questions require information found in the peptide, channels, or x-ray crystallography links.

1. The molecular models of the 3 amino-acid residues lining the channels of ferritin (shown in the peptide and residue links) show that all three of the residues contain polar groups. This seems to lead to a contradiction since one channel is hydrophilic (lined with polar groups) and the other channel is hydrophobic (lined with nonpolar groups). How can this occur?

2. A segment of a protein is analyzed and found to contain the amino-acid sequence Tyr-Asn-Val. The ChemDraw structures of these three amino acids are shown below.

a. Draw a two-dimensional sketch (Lewis structure) of this amino-acid sequence (like Peptide Figure 2).

b. If these 3 residues formed part of a channel, would the channel be hydrophobic or hydrophilic? Briefly, explain your reasoning.

3. Which representation used in the peptide link most clearly shows whether a protein contains alpha-helical regions or beta-pleated sheets?

4. Which representation used in the peptide link most clearly shows that ferritin is a hollow shpere (i.e., gives the most accurate overall shape of the molecule)?

5. Transferrin is another iron-carrying protein that functions in recycling the iron from red blood cells after they have died. Use Jmol to view the structure of the transferrin protein interactively.In Jmol, choose the most appropriate representation(s) from the "Display" menu to view the transferrin protein and answer the following questions about its structure:

a. Does transferrin contain alpha-helical regions? Which representation most clearly displayed the presence or absense of the helices?

b. Does transferrin consist of a hollow sphere for storing iron like the ferritin protein? Which representation most clearly displayed the presence or absense of the hollow sphere?

Mechanism of in Vitro Iron-Removal Process

In vitro studies on the removal of iron from ferritin are of interest because of ferritin's possible use in the diagnosis of iron overload and in increasing our understanding of iron metabolism. Below are two schemes that outline the in vitro iron-removal mechanism, as performed in the experiment Part II.B.

  1. Scheme 1: The first step in the reaction is the reduction of Fe(III) in the iron-mineral core to Fe(II) using Dihydroxyfumarate (DHF) as the reducing agent.
  2. Scheme 2: After Fe(III) is reduced to Fe(II), Fe(II) leaves the protein via the 3-fold channel (hydrophilic). Outside the protein, Fe(H2O) 62+ reacts with three (3) dianionic ferrozine ligands to form [Fe(ferrozine)3]4- .

Scheme 1:

This is a two-dimensional representation of the first step of the reaction. The iron-mineral core is attached to the ferritin shell. The protein wall is 10 Angstroms thick and is denoted by the double white line. The diameter of the inside of the protein shell is 80 Angstroms and the channels are represented by breaks in the protein shell. DHF may enter through a channel or may transfer an electron to the mineral core via another mechanism.

Note: This scheme is not drawn to scale.

Note: The carbon atoms are green, the oxygen atoms are red, the hydrogens are white, and the iron atoms are magenta in these stick representations.

Scheme 2:

This is a two-dimensional representation of the remaining steps in the reaction. After Fe(III) is reduced to Fe(II), Fe(II) leaves the protein via a 3-fold channel and reacts with three (3) dianionic ferrozine ligands outside the protein shell. The reaction of Fe(II) with ferrozine2- results in formation of the [Fe(ferrozine)3]4- compound.

Note: This scheme is not drawn to scale.

Note: The carbon atoms are green, the hydrogens are white, the iron atoms are magenta, the nitrogen atoms are blue, the oxygen atoms are red, and the sulfur atoms are yellow in these stick representations.

Questions on the Mechanism of the Iron-Removal Process

6. The stick representaion (e.g., the representations in Schemes 1 and 2) give a good picture of how the atoms in a molecule are connected to one another, but it does a poor job conveying some of the information provided by the CPK model (e.g., the representation in Figure 1).

a. What additional information could have been displayed if CPK representations had been used instead of stick representations in Schemes 1 and 2?

b. How would these schemes need to be modified (in addition to changing to the CPK representation) in order to accurately depict this additional information?

7. How many moles of [FeO(OH)]8[FeO(H2PO4)] would be present in a saturated ferritin protein (i.e., containing 4500 Fe atoms)?

8. Approximately 25% of the iron in the body is stored in ferritin, and 50% is found in hemoglobin. If ferritin in the body is not saturated, but typically contains only about 2000 iron atoms on average, calculate the ratio of moles of hemoglobin in the body to moles of ferritin in the body. (HINT: Refer to the Hemoglobin tutorial for the number of iron atoms per hemoglobin protein.)

9. What type of ligand is ferrozine? (Recall the ligands described in the Hemoglobin tutorial.)

10. Use Jmol to view the Fe(ferrozine)34- complex interactively. Rotate the molecule to obtain the best orientation for viewing the coordination of the central iron atom. (You may also wish to change the "Display" and "Color" settings to enhance your view of the atoms bonded to iron.)

a. Print the Jmol view that best shows the coordination of the central iron atom, and label the iron atom on your printout. Attach this printout to the tutorial assignment to be handed in.

b. What is the approximate geometrical shape about Fe in the Fe(ferrozine)34- complex? (Refer to the lab manual, Figure 1 in Experiment 3, for examples of the shapes of metal complexes.)

Compound-Structure Index

Figure 2 contains all of the main species involved in the in vitro iron-removal process described in this tutorial. This figure allows you to compare the relative sizes of the different systems involved. Clicking on each of the species in the picture also links you to a more detailed picture of the respective compounds. When you are looking at the species in detail, you should be predicting what the geometric structure would look like based on your intuition and how that structure affects the role that each species plays in the iron-removal mechanism.

Figure 2:

This is a molecular model representation depicting the relative sizes of the main species involved in the reaction. To see a more detailed picture of each species, please click on each of the species in the picture. Starting at the top of the figure on the right-hand side are Dihydroxyfumarate, Fe(II) hydrated complex, Ferrozine 2- ligand, and Fe(ferrozine)34-. The iron-mineral core is shown inside the protein shell.

Jmol Files

To view the molecules interactively, please use Jmol. To download the pdb files for viewing and rotating the compounds used in this experiment, please right-click on the appropriate compound name below, and select 'Save Link As' from the menu.

Ferritin and components of the ferritin protein:

Amino-acid residues:

Other species used in the experiment:

Other iron-carrying proteins:


Harrison, P.M.; Andrews, S.C.; Artymiuk, P.J.; Ford, G.C.; Guest, J.R.; Hirzmann, J.; Lawson, D.M.; Livingstone, J.C.; Smith, J.M.A.; Treffry, A.; Yewdall, S.J. in Iron Transport and Storage; Ponka, P.; Schulman, H.M.; Woodworth, R.C., Eds.; CRC: Boca Raton, FL, 1990; pp 81-101.

Theil, E.C. Annu. Rev. Biochem.1987, 56, 289-316.

Lawson, D.M.; Artymiuk, P.J.; Yewdall, S.J.; Smith, J.M.A.; Livingstone, J.C.; Treffry, A.; Luzzago, A.; Levi, S.; Arosio, P.; Cesareni, G.; Thomas, C.D.; Shaw, W.V.; Harrison, P.M. Nature 1991, 349, 541-544.


The development of this experiment was supported by the National Science Foundation Division of Undergraduate Education Grant# NSF/DUE-9455918 to the ChemLinks program and by a grant from the Howard Hughes Medical Institute through the Undergraduate Biological Sciences Education program, Grant HHMI# 71195-502005 to Washington University. The modifications to the tutorial were supported by a grant from the Howard Hughes Medical Institute through the Undergraduate Biological Sciences Education program, Grant HHMI#71199-502008 to Washington University. The molecular-modeling calculations were performed using the Washington University Department of Chemistry Computer Facility, which was funded in part by Washington University's Parents' Fund.

Copyright 1995, Washington University, All Rights Reserved.

[To Ferritin web page]

Revised 10/1/98