The C-Terminal Domain

The G-terminal portion of glutamate receptor subunits is the most diverse region of the protein, ranging from short sequences of less than 50 amino acids to the large signaling complexes of NMDA R2 subunits. In contrast to the S1S2 domain, the structures of the G-terminal domains are less well understood. The relatively short G-terminal domains of AMPA and kainate receptors do not adopt a stable structure when expressed as 

Relating Structure, Function, and Dynamics Channel Gating and Desensitization 

I gale open 3 gjiics dosed litlle OT no conduelancc 

1 Open gate Parlially closed lobe Fully dosed lobe 

---

Figure 6.20 Space-filling diagram illustrating the structural changes of CDK2 upon cyclin binding, (a) The active site is in a cleft between the N-terminal domain (blue) and the C-terminal domain (purple). In the inactive form this site is blocked by the T-loop.

In spite of the absence of the C-terminal domains, the DNA-binding domains of lambda repressor form dimers in the crystals, as a result of interactions between the C-terminal helix number 5 of the two subunits that are somewhat analogous to the interactions of the C-terminal p strand 3 in the Cro protein (Figure 8.7). The two helices pack against each other in the normal way with an inclination of 20° between the helical axes. The structure of the C-terminal domain, which is responsible for the main subunit interactions in the intact repressor, remains unknown. 

Figure 8.7 The N-terminal domains of lambda repressor form dimers, in spite of the absence of the C-terminal domains that are mainly responsible for dimer formation in the intact repressor. The dimers are formed by interactions between a helix 5 from each subunit. The different subunits are colored green and brown, except the helix-turn-hellx motif, which is colored blue and red as in Figure 8.4. (Adapted from C. Pabo and M. Lewis, Nature 298 443-447, 1982.)...

Many biochemical and biophysical studies of CAP-DNA complexes in solution have demonstrated that CAP induces a sharp bend in DNA upon binding. This was confirmed when the group of Thomas Steitz at Yale University determined the crystal structure of cyclic AMP-DNA complex to 3 A resolution. The CAP molecule comprises two identical polypeptide chains of 209 amino acid residues (Figure 8.24). Each chain is folded into two domains that have separate functions (Figure 8.24b). The larger N-terminal domain binds the allosteric effector molecule, cyclic AMP, and provides all the subunit interactions that form the dimer. The C-terminal domain contains the helix-tum-helix motif that binds DNA. 

TBP mutants lacking the N-terminal region are fully functional in promoter binding and stimulation of basal transcription and therefore these two functions must be provided by the C-terminal domain. Furthermore, the C-terminal domain of yeast TBP contains all the functions essential for normal yeast cell growth and for responses to specific transcriptional activators with a net negative charge. This C-terminal domain contains two homologous... 

Figure 9.12 Schematic diagram of the structure of the heterodimeric yeast transcription factor Mat a2-Mat al bound to DNA. Both Mat o2 and Mat al are homeodomains containing the helix-turn-helix motif. The first helix in this motif is colored blue and the second, the recognition helix, is red. (a) The assumed structure of the Mat al homeodomain in the absence of DNA, based on Its sequence similarity to other homeodomains of known structure, (b) The structure of the Mat o2 homeodomain. The C-terminal tail (dotted) is flexible in the monomer and has no defined structure, (c) The structure of the Mat a 1-Mat a2-DNA complex. The C-terminal domain of Mat a2 (yellow) folds into an a helix (4) in the complex and interacts with the first two helices of Mat a2, to form a heterodimer that binds to DNA. (Adapted from B.J. Andrews and M.S. Donoviel, Science 270 251-253, 1995.)...

The polypeptide chain of p53 is divided in three domains, each with its own function (Figure 9.16). Like many other transcription factors, p53 has an N-terminal activation domain followed by a DNA-binding domain, while the C-terminal 100 residues form an oligomerization domain involved in the formation of the p53 tetramers. Mutants lacking the C-terminal domain do not form tetramers, but the monomeric mutant molecules retain their sequence-specific DNA-binding properties in vitro. 

The C-terminal domain of phosducin is a five-stranded mixed p sheet with a helices on both sides, similar to the thioredoxin fold of disulfide iso-merase DsbA described in Chapter 6. Despite significant sequence homology to thioredoxin, the phosducin domain, unlike other members of this family. 

A unique feature of the interaction of the hormone and PLR is at the beginning of the F-G loop in the C-terminal domain. In HGR the sequence is Arg-Asn-Ser whereas in PLR it is Asp-His-deletion. This loop interacts with His 18 and Glu 174 of the hormone. In PLR the orientation of this loop is such that the Asp and His residues, in combination with His and Glu from the hormone, form a strong binding site for a zinc atom that links the hormone and the receptor (Figure 13.23b). The presence of zinc increases the affinity of the hormone for the receptor in vitro by a factor of 10,000. As shown by mutagenesis studies His 18 and Glu 174 of the hormone are important for the tight binding to PRL but not to GHR. 

Figure 13.23 The F-G loop in the C-terminal domain of the prolactin receptor is involved in a unique interaction, (a) The F-G loop of the growth hormone receptor (blue) is not involved in any specific interactions with the growth hormone (red), (b) The F-G loop in the prolactin receptor forms a strong zinc-binding site that links the receptor (green) to the hormone (red). (Adapted from W. Somers et at.. Nature 372 478-481, 1994.)...

Figure 17.3 The polypeptide chain of lysozyme fiom hacteiiophage T4 folds into two domains. The N-terminal domain is of the a + P type, built up from two a helices (red) and a four-stranded antiparallel P sheet (green). The C-terminal domain comprises seven short a helices (brown and blue) in a rather irregular arrangement. (The last half of this domain is colored blue for clarity.)...

Figure 18.16 One-dlmenslonal NMR spectra, (a) H-NMR spectrum of ethanol. The NMR signals (chemical shifts) for all the hydrogen atoms In this small molecule are clearly separated from each other. In this spectrum the signal from the CH3 protons Is split Into three peaks and that from the CH2 protons Into four peaks close to each other, due to the experimental conditions, (b) H-NMR spectrum of a small protein, the C-terminal domain of a cellulase, comprising 36 amino acid residues. The NMR signals from many individual hydrogen atoms overlap and peaks are obtained that comprise signals from many hydrogen atoms. (Courtesy of Per Kraulis, Uppsala, from data published in Kraulis et al.. Biochemistry 28 7241-7257, 1989.)...

Figure 18.17 Two-dimensional NMR spectnim of the C-terminal domain of a cellulase. The peaks along the diagonal correspond to the spectrum shown in Figure 18.16b. The off-diagonal peaks in this NOE spectrum represent interactions between hydrogen atoms that are closer than 5 A to each other in space. From such a spectrum one can obtain information on both the secondary and tertiary structures of the protein. (Courtesy of Per Kraulis, Uppsala.)...

Figure 18.20 The two-dimensional NMR spectrum shown in Figure 18.17 was used to derive a number of distance constraints for different hydrogen atoms along the polypeptide chain of the C-terminal domain of a cellulase. The diagram shows 10 superimposed structures that all satisfy the distance constraints equally well. These structures are all quite similar since a large number of constraints were experimentally obtained. (Courtesy of P. Kraulis, Uppsala, from data published in P. Kraulis et ah. Biochemistry 28 7241-7257, 1989, by copyright permission of the American Chemical Society.)...

The a subunits, for which two isoforms exist in mammals (al, a2), contain conventional protein serine/threonine kinase domains at the N-terminus, with a threonine residue in the activation loop (Thr-172) that must be phosphorylated by upstream kinases (see below) before the kinase is active. The kinase domain is followed by an autoinhibitory domain, whose effect is somehow relieved by interaction with the other subunits. The C-terminal domain of the a subunit is required for the formation of a complex with the C-terminal domain of the (3 subunit, which in turn mediates binding to the y subunit. The al and a2 catalytic subunit isoforms are widely distributed, although a2 is most abundant in muscle and may be absent in cells of the endothelial/hemopoietic lineage. 

The exchange of cations for protons occurs through the transmembrane domain, whereas regulation (fine-tuning) of the exchanger is largely exerted by the C-terminal domain. NHEs are believed to form homodimers in the membrane. 

Phosphorylation of HSF substantially enhances the transcriptional activity of HS gene expression which may be up to 100-fold of basal levels after HSFl binds to the promoter element. Heat shock will increase the C-terminal-domain-kinase activity in cell extracts, and this action may enhance the activity of RNA polymerase II that is bound to HS genes (Legagneux et al., 1990). Whether this kinase activity also affects HSFl phosphorylation is not known, but increased HS gene expression appears to occur as long as HSFl is bound to the promoter region. The CTD kinase complex contains multiple proteins, and it is quite possible that one or more of these proteins is also regulated by stress. 

HIV integrase consists of three distinct domains. The N-terminal domain contains a HHCC motif that coordinates a zinc atom that is required for viral cDNA integration. Three highly conserved amino acids (D,D-35-E) are embedded in the core domain, which form the acidic catalytic triad coordinating one or possibly two divalent metals (Mn + or Mg +). The C-terminal domain (residues 213-288) is responsible for unspecific DNA binding and adopts an overall SH3 fold (Chiu and Davies 2004). The enzyme functions as a multimer and to this end all three domains can form homodimers. 

Fig. 2). Not all MMPs have these domains, for example, matrilysin lacks the C-terminal domain and the MT-MMPs do not have the propeptide. Some MMPs have additional domains, for example, the gelatinases have three repeats of a fibronectin-type II domain. 

Figure 5.9. Split ubiquitin as a sensor for protein-protein interactions. Protein A is fused to the N-terminal domain and protein B is fused to the C-terminal domain of ubiquitin. Interaction of A and B reconstitutes a full-sized, folded ubiqutin. The folded ubiquitin is recognized by a specific protease and cleavage releases the reporter protein.

Riechmann, L., and Holliger, P. (1997). The C-terminal domain ofTolA is the coreceptor for filamentous phage infection ofE. coli. Cell 90, 351-360. 

The N-terminal domain of the OCP is an orthogonal alpha-helical bundle, subdivided into two four-helix bundles (Figure 1.3a and c). These subdomains are composed of discontinuous segments of the polypeptide chain (gray and white in Figure 1.3c). To date, the OCP N-terminal domain is the only known protein structure with this particular fold (Pfam 09150). The hydroxyl terminus of the 3 -hydroxyechinenone is nestled between the two bundles. The C-terminal domain (dark... 

---