CRP

cAMP Receptor Protein (CRP) is a dimer of two identical subunits each of which is 209 amino acids in length. Crystal structures have been solved for:

 

• The CRP-cAMP complex at 2.9Å and 2.5Å
  
  
• The CRP-cAMP-DNA complex at 3.0Å and 2.5Å
  

A crystal structure for the CRP protein dimer by itself has not yet been solved.

The following image from from the ExPASy Molecular Biology Server (Canadian Mirror) shows the structure of CRP-cAMP bound to DNA:

Note that CRP bends DNA when it binds. Also note that the DNA binding site in this structure consists of two DNA molecules (coloured green and blue). You can see that the break in these two molecules is located very close to the polypeptide backbone of CRP. This was the major reason why Parkinson et al. (1996) recrystallized the complex using slightly different DNA molecules so that the contacts in this region could be established correctly.

 

• Each monomer of the protein has two structural and functional domains. The N-terminal domain (amino acids 1 - 140, approx.) contains the cAMP nucleotide binding site. The C-terminal 50-60 amino acid domain contains a helix-turn-helix DNA binding site.
  
  
• Each monomer can bind to a single molecule of cAMP. There is still some question as to whether the active form of the protein in vivo contains one or two bound cAMP molecules. Many studies indicate that CRP is fully active with just a single bound cAMP and that this is the most likely form of the active molecule in the cell.
  
  
• The bound cAMP has an anti- glycosidic bond conformation in the crystal structures.
  
  
• The binding of cAMP changes the conformation of the protein so that the DNA binding alpha helices are brought into the correct alignment to the able to recognize and bind to the binding site. In the crystal structure, cAMP is indirectly involved in bonds that link the 2 monomers in a dimer as well as in bonds that link the two domains within a monomer. Thus cAMP binding alters the internal bonding patterns in the dimer.

See also EBI.

 

CRP binding to DNA

GENERAL FEATURES

The following are some general features of the CRP binding site:

 

• CRP-cAMP will bind to nonspecific DNA as well as to its specific DNA binding site. By measuring the binding constant as a function of salt concentration, we can estimate that nonspecific binding involves 6 electrostatic interactions between the protein and DNA while specific binding involves 8 electrostatic interactions (as well as specific base-pair recognition hydrogen bonds).
  
  
• DNase digestion of CRP-DNA complexes established that CRP bound to a 26 bp region of DNA.
  
  
• Nucleotide sequencing and analysis of CRP binding sites established a consensus binding sequence consisting of an imperfect 5 bp palindrome:
  
  
  

  TGTGA --- N6 --- TCACA

  
  
  
• Genetic and crystallographic evidence suggests, however, that functionally significant nucleotides extend 11 bp on either side of axis of symmetry.
  
  
• To explain the ionic strength dependence of binding as well as to account for all of the contacts made between CRP and its binding site, 14 bp on either side of axis of symmetry are required.
  

Note, however, that no real CRP site has two identical perfect half sites. In fact, the binding sites for most DNA-binding proteins have this characteristic. It seems that perfect binding sites would be too strong for functional usefulness in the cell. Thus all known CRP binding sites have different half site sequences. Each site will bind CRP with different affinity but one that is appropriate for its function at that site. Artificial CRP binding sites in which both halves consist of perfect consensus sequences have been created. They bind CRP 50x stronger than the lac operon CRP binding site does. The lac operon CRP binding site is one of the stronger CRP binding sites in the E. coli cell; other CRP binding sites are up to 50 fold weaker.

 

SPECIFIC FEATURES

Specific features of the CRP binding site have been determined from genetic analysis, methylation and ethylation protection experiments and from the co-crystal complex.

 

• The L8 and L29 mutations are symmetric mutations which change the same base pair on each side of the axis of symmetry within the conserved 5 bp sequence:
  

 

The L8 mutation reduces the affinity of CRP for the binding site by a factor of ten.

• Methylation protection experiments show that methylation of two GUANINES in each half-site -- also within the 5 bp consensus sequence -- are affected by the presence of CRP.
  
  

 

Methylation of two GUANINES within the central nonconserved core is enhanced by the presence of CRP.

• The pattern of ethylation protection indicates that CRP binds to one face of a DNA double helix. Each cluster of protected phosphates is centred 5 bp from the next. They appear to be on different strands. However, since the periodicity of DNA is 10 bp, each cluster will be located on the same side of a three dimensional DNA double helix.
  
  
• The cocrystal between CRP-cAMP and DNA has confirmed much of the above data and established the identities of the various amino acids in CRP that are responsible for particular contacts or interactions:
  

Image from Parkinson, G., Wilson, C., Gunasekera, A., Ebright, Y.W., Ebright, R.E. & Berman, H.M. (1996) Structure of the CAP_DNA Complex at 2.5Å Resolution: A Complete Picture of the protein-DNA Interface. J. Mol. Biol. 260: 395-408.

 

 

CRP interacts with 24 out of 30 bp in the DNA fragment used in the cocrystals.

All of the interactions between the DNA and the protein involve amino acids in the C-terminal domain -- except for one: Lysine 26 in one of the two subunits interacts with the DNA binding site.

The N-terminal end of alpha helix F is parallel to the plane of the base pairs and penetrates the major groove of the DNA

The majority of the interactions between the protein and the DNA site are "weaker" interactions with the phosphate backbone. (K26, K166, R169, Q170, S179, T182, H199). Some of these interactions are not with amino acid side chains but with the amide hydrogen of the peptide bond between two amino acids. (169NH, 170NH, 179NH)

Only three amino acids make specific contacts with base pairs:

• R180 is hydrogen bonded with the N7 and O6 of GUANINE 5
  
  
• E181 is hydrogen bonded with the N4 of CYTOSINE 7'
  
  
• R185 is hydrogen bonded with the N7 and O6 of GUANINE 7 and also with water molecules that form hydrogen bonds to a phosphate and to GUANINE 9'
  
  

Finally, when CRP binds to DNA it makes two distinct bends in each half site. There is a sharp kink of 40° between positions 6 & 7. This explains why GUANINE 7 shows enhanced methylation when CRP is bound to DNA. A smaller bend is also found between positions -1 and 1 of each half site. (Refer to the figure/handout for the numbering system)

DNA bending -- whether naturally occurring or caused by the binding of proteins to DNA -- can be observed by gel electrophoresis. DNA fragments in which a bend is located in the middle of the fragment migrate more slowly than otherwise identical fragments in which the bend is located near the end of a fragment.

In order to generate suitable DNA fragments to test whether DNA bending occurs, all one has to do is to clone two fragments next to one another. Then it by using the appropriate restriction enzymes one can generate identical but linearly permuted fragments in which the position of a bend will vary.

CRP @ Kenyon.