Solvent-exposed loops

Kallikreins can be roughly divided into two categories, the classical kallikreins (hKl, hK2, and hK3) and the new kallikreins. The new kallikreins appear to be unique in their three-dimensional structure and share some features with trypsins and other features with the classical kallikreins. Comparative protein models show that the pattern of hydrophobic side-chain packing in the protein core is nearly identical in all human kallikreins, and the observed differences occur within the solvent-exposed loop segments. 

In the given examples of native state isomerization for ITK and MS2 the trans form corresponds to an extended, solvent-exposed loop conformation. Cis forms are characterized by extensive contacts between the loops and the surface of the proteins compared to the trans conformers, where those contacts are weak or not present. It is speculated that such contacts provide the necessary energy needed to stabilize the inherently less stable cis conformation [147]. 

Since the outside of the barrel faces hydrophobic lipids of the membrane and the inside forms the solvent-exposed channel, one would expect the P strands to contain alternating hydrophobic and hydrophilic side chains. This requirement is not strict, however, because internal residues can be hydrophobic if they are in contact with hydrophobic residues from loop regions. The prediction of transmembrane p strands from amino acid sequences is therefore more difficult and less reliable than the prediction of transmembrane a helices. 

The authors expressed PKA consisting of 353 amino acids, of which eight are prolines. Resonances of 274 backbone amide peaks were visible in the spectrum, of which 191 were assigned. It was possible to assign resonances for the N- and C-terminal sequences, the majority of the N-lobe, including the glycine-rich loop, and most of the solvent-exposed residues of the C-lobe. This enabled a determination of the structure for the more flexible parts of the structure. However, many correlations were missing for the... 

The five N-terminal residues and the six or seven C-terminal residues cannot be seen in the high resolution electron density map, and the loop referred to above, formed by residues 44 to 53, appears at only one-third to one-half the amplitude of the well-resolved parts of the map. The lack of clarity in these three regions might possibly result from poor phasing or some other crystallographic factor, but we consider it more likely that these predominantly hydrophilic sections of the peptide project in a disordered way into the solvent. In this connection, it is interesting that in the presence of Ca2+ and pdTp trypsin cleaves inhibited nuclease at only two points between residues 5 and 6 and between residues 48 and 49 (36-38) which are at the very extremity of the loop. It also seems relevant that ribonuclease S also shows lack of clarity at the ends of the peptide chains and in the region of a relatively exposed loop (56). 

The complex with ADP and aluminium fluoride is thought to resemble the early ADP.Pi state immediately after hydrolysis, whereas the complex with ADP and vanadate may represent a late state in which the phosphate (mimicked by vanadate) has moved quite a long distance (15 A) from the active center to the surface of the motor domain. There it is fixed by two hydrogen bonds to the solvent exposed tips of the switch-1 loop region (L9) at one side and the switch-2 loop (Lll) at the other side. 

Fig. 3.12. Ribbon structure (A) and L2 loop (B) of EcCM. Random mutagenesis of the L2 loop followed by selection for chorismate mu-tase activity in vivo showed little sequence constraint on solvent exposed turn residues, aside from a modest bias in favor of hydrophilic amino acids [85]. In contrast, long-range tertiary contacts impose a strict requirement for hydrophobic aliphatic amino acids at position 68.

Fig. 16. Plots of e° versus sequence for the solvent-exposed sites in each of the loops investigated. The approximate position of the membrane-aqueous interface relative to the rhodopsin sequence is located at the transitions between white and grey background.

Fig. 14. X-ray crystal structure of full-length yeast CCS [pdb code Iqup (Lamb et al., 1999)]. (a) One monomer of yCCS is in light gray and the other is in dark gray. The cysteine residues of the MXCXXC motif in domain 1 are labeled and form a disulfide bond in each subunit. Amino acid side chains that are important in the formation of the positive patch at the dimer interface (Arg-188 and Arg-217) and the solvent-exposed Trp-183 residues of loop 6 at the center of this patch are shown in ball-and-stick representation. Domain 3 is not visible in the crystal structure (see text), (b) Stereo view of the image in (a) rotated 90° in the horizontal plane of the page and then 90° counterclockwise around an axis perpendicular to the page. The side chains that form the putative ySODl interaction surface are represented as ball-and-stick. The cysteine residues of the domain 1 MXCXXC motif are also represented in ball-and-stick.

Solvent is usually excluded from the blue copper site, which is buried 6 A inside the protein, having only the His ligand from the copperbinding loop exposed to the surface. The phytocyanins, stellacyanin and plantacyanin (cucumber basic protein), are exceptions, in which both His ligands are solvent exposed and the copper ion is only 3 A beneath the protein surface. This situation makes the copper center in this family of blue copper proteins more accessible to low-molecular-weight solutes (see Section V). 

---