Graphics Gallery

Gale Rhodes
Chemistry Department
University of Southern Maine

Revised 2006/11/07

Learn how to use Swiss-PdbViewer. Work through sections 1-4 of the Swiss-PdbViewer Tutorial.

Topic: Bioenergetics

Studies of proteins commonly discussed in biochemistry texts

Examples:

Adenylate Kinase

This enzyme catalyzes the reaction AMP + ATP <==> 2 ADP. Read about the adenylate kinase reaction in your textbook.

Click here to download a Swiss-PdbViewer project file comparing E. coli adenylate kinase with (holo, 1AKE) and without (apo, 4AKE) the bisubstrate analog Ap5A. Ap5A is like the two substrates, AMP and ATP (or two products, ADP and ADP), joined by a fifth phosphoryl group. (The models were superimposed with DeepView using Fit:Magic Fit.)

The project file contains four layers -- two cycles of 4AKE (holo) alternating with 1AKE (apo). In the first cycle (layers 1 and 2), the models are colored CPK. In the second cycle (layers 3 and 4) the models are colored by B-factor, which in high quality model reflects relative mobility of the residues.

Conformational Changes Upon Ligand Binding

First, in the Layer Infos window, in the cyc column, check to put "-" for layers 3 and 4, and click to put check marks for layers 1 and 2. Observe the differences between the models by blinking between them (hold down control and press tab repeatedly). Notice that the enzyme almost completely envelops the bisubstrate analog.

  • Describe the differences conformations of the two models. Identify and describe portions of the structure that move relative to the core pleated sheet. Make this easier by displaying the models as ribbons.
  • Color the second layer by RMS to emphasize the conformational changes.
  • Display residues within 5 angstroms of the analog. What types of interactions (H-bonds, ionic attractions, hydrophobic interactions) bind the substrates to the enzyme?
  • Why might this enzymatic reaction require protecting the substrate completely from water?

Mobility Changes Upon Ligand Binding.

In the Layer Infos window, click to put "-" for layers 1 and 2, and to put checkmarks for layers 3 and 4, which show both models colored by B-factors. In models of high quality and resolution, B-factors show relative mobility: blue residues are the most rigid ("cold"), while red ones are the most mobile ("hot").

This enzyme appears to bind its substrate very tightly, yet enzymes must readily release products to be catalytically efficient. Here's how one textbook author describes this change: "On binding substrate, a portion of the protein remote from the active site increases its chain mobility and thereby consumes some of the free energy of substrate binding. The region 'resolidifies' when the binding site is opened and the products are released. This mechanism is thought to act as an 'energetic counterweight' to help adenylate kinase maintain a high reaction rate."

  • Use B-factor color to help you find the region that provides this energetic counterbalance to substrate binding. Does this model appear to support the contentions of the text?
  • You should find the text's explanation puzzling. After all, increased disorder is energetically favorable (DG = DH - TDS), and would appear to make the overall DG even more negative. Let's analyze this effect more carefully. The text is saying that DG of substrate binding is negative (substrate binding is spontaneous), but lower in magnitude than expected for such extensive binding, because the negative DG of substrate binding is partially counterbalanced by the positive DG of changes in the remote region. This remote region, we are told, becomes more mobile -- that is, more disordered. Consider these questions:
  1. Would increased disorder in the remote region upon substrate binding make a positive or negative contribution to the overall DG of substrate binding?
  2. More disorder implies fewer noncovalent interactions in the remote region. Would the breakage of noncovalent interactions make a positive or negative contribution to the overall DG?
  3. Finally, in the light of questions 1 and 2, is the contribution of the remote region primarily entropic or enthalpic? In other words, is the free energy change of substrate binding counterbalanced by mobility (entropy) or by broken bonds (enthalpy) in the remote region?

Answers to Questions

First, the model appears to contradict the contentions of the text. According to the B-factors, a large region is more mobile when the substrate analog is absent rather than when it is present. Locate TYR133 in the apo model. It sits in the middle of the region I am describing. Notice that the B-factor colors in this wing of the protein are mostly yellow, orange, and red, indicating that it is quite disordered (that is, if this is a high quality model and B-factors are really telling us something about mobility). Notice that B-factor colors in this region are mostly blue-green in the holo model, indicating that the region became more ordered. So the models do not agree with the textbook description.

Now to the numbered questions:

  1. The disordering of a region due to substrate binding would make DG of binding even more negative, thus increasing the tenacity of binding, and making release of products slower. Increased order due to substrate binding would make DG less negative and thus reduce the tenacity of binding, making release of products faster. This appears to be exactly what is happening.
  2. Breakage of noncovalent bonds requires energy (DH > 0), so makes DG more positive. Breakage of noncovalent bonds due to substrate binding would make binding less tenacious, and make release of products faster. Formation of covalent bonds due to substrate binding would make DG of binding more negative, and make release of products slower. If a remote region of the protein were actually becoming more disordered due to substrate binding, the entropy increase would favor binding, but the breakage of noncovalent bonds would work to weaken binding.
  3. Remember that when proteins unfold, even partially, internal hydrogen bonds probably get replaced by hydrogen bonds to water. So the net DH is probably near zero. The effect of "melting" a region of the protein is probably primarily entropic.

I conclude that the textbook explanation is incorrect. The remote region under discussion is disordered in the absence of the substrate analog, and it becomes quite highly ordered when the analog binds (judging from the B-factors). If the main energy difference between the two conformations of the region that contains TYR133 is entropic and not enthalpic, then the ordering of this region upon substrate binding makes DG of binding less negative, and reduces tenacity of binding. This would lower the energy barrier to release of products and make the reaction faster, in keeping with the effect the author is trying to explain.

Topics List

Biochemistry Resources