Protein Engineering, Vol. 15, No. 8, 697-705,
August 2002
© 2002 Oxford University Press
Conserved water molecules in MHC class-I molecules and their putative structural and functional roles
1 Service de Conformation de Macromolécules Biologiques, Centre de Biologie Structurale et Bioinformatique, Université Libre de Bruxelles,av. F.D. Roosevelt 50, CP160/16, B-1050 Brussels, Belgium and 2 European Bioinformatics Institute (EBI), Genome Campus, Hinxton, Cambridge CB10 1SD, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
A set of conserved water positions making direct contacts with the
1 and 2 domains of the MHC class-I protein was identified by a cluster analysis in 12 high-resolution crystal structures of proteins from different allele types and different species, comprising human, mouse and rat. The analysis revealed a total of 63 clusters, corresponding to water molecules, whose positions are conserved in half or more of the analyzed structures. Analysis of these clusters shows that the most conserved water positions—those appearing in the largest fraction of the structures—were also the most accurately defined, as measured by their normalized crystallographic B-factor. Not too surprisingly, these positions displayed better overlap and formed more H-bonds with the protein. In a second part of this work, a detailed analysis is presented of three of the most conserved water positions and their putative structural and functional roles are discussed. The most highly conserved of the three appears to play an important role in stabilizing the conformation of a twisted ß-turn between residues 118 and 122 (numbering of HLA-B3501, PDB code 1A1N). An equivalent water molecule was found to be associated with a similar ß-turn in 43 unrelated structures surveyed in the PDB, leading to the suggestion that this water molecule plays an important structural role in this type of turn. The second water molecule makes hydrogen bonds with residues lining pocket B in the peptide-binding groove and is suggested to play a role in modulating peptide recognition. The third highly conserved water molecule is located at the first kink of the 2 helix, possibly playing a role in determining the position of the N-terminal segment of that helix, which also carries side chains in contact with the bound peptide. This information on conserved water positions in MHC class-I molecules should be helpful in modeling interactions with bound peptide antigens and in designing new peptides with tailor-made affinities.
Keywords: B-factor/cluster analysis/conserved water molecules/MHC class-I
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
The major histocompatibility (MHC) class-I and class-II proteins play a central role in the immune response. They form complexes with peptide fragments of intact proteins and present them at the cell surface. The recognition of these complexes by the antigen-specific T cell receptors (TCRs) initiates the cellular immune response and is the basis whereby the cellular immune system discriminates between self and non-self [for reviews, see Rothbard and Gefter (Rothbard and Gefter, 1991
) and Sewell et al. (Sewell et al., 2000)].
The tertiary structures of an increasing number of MHC-class-I (Bjorkman et al., 1987a,b; Madden et al., 1992; Collins et al., 1995; Wang et al., 1995; Reid et al., 1996; Smith et al., 1996a,b; Glithero et al., 1999; Speir et al., 1999, 2001; Tormo et al., 1999; Maenaka et al., 2000; Hillig et al., 2001) and class-II (Stern et al., 1994; Ghosh et al., 1995; Fremont et al., 1996, 1998; Murthy and Stern, 1997; Scott et al., 1998) molecules have been determined by X-ray diffraction. The October 2001 release of the Protein Data Bank (RCSB-PDB) (Berman et al., 2000), contains 68 entries for MHC class-I molecules, comprising nine allele types from human, four from mouse and one from rat. Valuable information on the atomic interactions between the MHC molecule and bound peptide antigens and on those in the ternary complex of MHC with the peptide and the T-cell receptor molecules has been obtained from the tertiary structures of the corresponding complexes (Garboczi et al., 1996). The peptide binding groove in the MHC class-I molecules has been described as featuring six pockets that interact with the side chains of the bound peptide and are lined by amino acid residues, which differ between the allele types (Bjorkman et al., 1987a; Saper et al., 1991; Matsumura et al., 1992). Following this description, these six pockets have been classified on the basis of their selective recognition of specific amino acid types in residues of the bound peptide (Zhang et al., 1998).
Water positions conserved in different crystal structures of the same protein, in particular completely buried positions that are isolated from the bulk solvent, have been analyzed by several authors (Rashin et al., 1986; Williams et al., 1994). Several studies were also devoted to the analysis of conserved water positions in crystal structures of related proteins such as the serine protease family (Sreenivasan and Axelsen, 1992) or the microbial RNases (Loris et al., 1999). These and other studies showed that conserved water molecules can occur in enzyme active sites, where they seem to play a role in ligand binding (see, for example, Kryger et al., 2000) or catalysis. They can also be found in polar cavities located inside proteins, where they are believed to enhance protein stability. Conserved water positions in protein crystal structures may therefore correspond to water molecules that play structurally or functionally important roles.
The interactions of water molecules with the MHC proteins have also received attention, primarily in relation to peptide binding. They have been analyzed in X-ray structures of MHC class-I peptide complexes (Madden et al., 1992; Fremont et al., 1995; Smith et al., 1996a) or in the context of molecular dynamics simulations of these complexes (Meng et al., 1997, 2000). These studies mainly focused on the variation in bound water in the peptide binding groove and its role in determining peptide-binding propensities.
In this work, we took the converse approach and analyzed water positions conserved across the crystal structure of MHC class-I–peptide complexes from a recent release of the PDB. To identify such positions we applied an automatic clustering procedure to water molecules in 11 high-resolution well refined MHC–peptide complexes, after superposition of the corresponding protein backbones. This led to the identification of a total of 67 clusters of water molecules. Correlations between various properties of these clusters, such as the number of water molecules in each cluster, their conservation across the different crystal structures, their normalized B-factors and the average distance between molecules in the cluster, were established. In addition, we identified three very highly conserved water molecules, located at key positions in the MHC protein. The particular roles that these molecules may play in maintaining the stability of the free and peptide-bound states of MHC are discussed.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
MHC class-I–peptide complexes
From the total of 68 entries for class-I MHC–peptide complexes in the October 2001 release of the PDB, 12 high-quality, high-resolution (better than 2.65 Å) complexes, with many identified water molecules were selected (Table I). The H-2Db allele has four entries in the PDB, 1BZ9, 1CE6, 1HOC and 1QLF. Of those, 1HOC has the highest resolution (2.4 Å), but since the number of water molecules in this entry is smaller than in 1QLF, a structure of somewhat lower resolution, the latter structure was used instead. We also included the murine H-2M3 molecule (RCSB-PDB code 1MHC) which is class-Ib and as such is not polymorphic: it binds only N-formylated peptides. Its conformation is, however, very similar to those of the other MHC molecules in our data set, therefore qualifying as a valid example in our study.
|
Identification of conserved water positions by cluster analysis
The first step in our approach was to transform the atomic coordinates of all 12 selected structures into the same reference frame. To that end, the backbones of the 1 and 2 domains of the MHC molecules were superimposed on to those of HLA-B3501 (1A1N). Next, water molecules distant by <5 Å from any protein atom of the 1 and 2 domains in each individual crystal structure were selected and their coordinates were transformed into the same reference frame. These coordinates were then subjected to the cluster analysis.
This analysis was performed using the Average Group method (Sneath and Sokal, 1973; Kendall, 1975), with the Euclidean distance between water molecules as the metric. The distance cutoff for defining clusters was set to 1.5 Å. This particular value was used to ensure that clusters do not contain several water molecules belonging to the same crystal structure, which may be within hydrogen boding distance of one another.
Clusters grouping water molecules from at last six crystal structures (conservation ratio 
0.5) were retained for further analysis and the average coordinate of the oxygen atoms in each cluster was computed.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
Conserved water positions in MHC molecules
Our cluster analysis identified a total of 67 clusters of conserved water positions in the 12 crystal structures of MHC class-I molecules listed in Table I
. Average coordinates of water molecules in each cluster and the members of the cluster from the different MHC molecules are listed in Table II.
|
An interesting aspect to consider is the relative contribution from MHC structures of different alleles and species to the conserved water positions identified here. Clusters 1–4, with the most highly conserved water positions, those with positions from at least 10 out of the 12 analyzed structures (conservation ratio 0.83–1.0), contain water molecules from all human MHC structures (Table II
). On the other hand, one classical (1KBG) and one non-classical (1MHC) murine molecules and one from rat (1ED3) contribute one and two water molecules, respectively, less to clusters in this conservation range (see Table II). These proteins also tend to contribute to fewer clusters than the other analyzed variants (bottom row of Table II). However, inspection of Tables I and II also indicates that there is a correlation between the total number of crystallographic water molecules in contact with the 1 and 2 domains in each MHC structure and the number of clusters (with a conservation ratio of 0.5) to which it contributes. Given that the conformations of the protein portion surrounding the conserved water molecules are very similar in all the analyzed structures, we may expect that the murine and rat structures (1KBG and 1ED3) would probably display these conserved water molecules in more accurate X-ray structures. A similar statement can be made for the human HLA-AW68 (1TMC). These observations suggest that the pattern in the number of highly conserved water molecules in the MHC class-I structures is probably not related to the species from which the molecules originate and may therefore be representative of MHC structures in general.
The positions of the identified clusters of conserved molecules are shown in Figure 1 together with the ribbon drawing of the reference MHC molecule (HLA-B3501, RCSB-PDB code, 1A1N). The views shown are from the directions of the 3-domain (Figure 1a) and the bound TCR (Figure 1b), respectively. The clusters are displayed as spheres whose radii are proportional to the average distance between water molecules in each cluster, representing a measure of the spatial spread of the cluster. The color of each sphere reflects how well the water positions in the cluster are conserved across the analyzed crystal structures.
|
Inspection of Figure 1
reveals that most of the clusters with conservation ratios >0.75 are located on the backside of the ß-sheet, which is facing the 3-domain (Figure 1a). This side also contains the cluster with the highest conservation ratio of 1, highlighted by a green circle. On the other hand, many of the clusters with less well conserved water positions and comparatively large radii are located on the opposite side, which contains the peptide binding pocket and faces the TCR receptor (Figure 1b). Interestingly, we see that clusters containing highly conserved water positions—those from nine or more structures—also tend to have smaller radii. The smallest average distance between water positions (0.35 Å) is observed for the cluster with the highest conservation ratio (1.0) amongst all 67 clusters. Analysis of the parent 3D structures shows that the water molecule whose positions are grouped in this tight cluster forms three hydrogen bonds with the nearby main chain atoms. Inspection of the superimposed backbones shows, moreover, that the main chain conformation in this region is well conserved in the 12 MHC molecules (see below).
In contrast, several other clusters in the vicinity of this tight cluster highlighted by a yellow circle in Figure 1a have a comparatively larger radius. Inspection of the superimposed MHC backbones in this region reveals that these display appreciable conformational variability. These more variable clusters are located near the beginning of the 2 helix. The backbone structure in this region also displays appreciable variability across the 12 MHC class-I molecules examined. The variability in the water positions of these clusters therefore probably relates to the variability in local backbone conformation.
The purple circle in Figure 1a highlights a cluster (No. 23) with a large dispersion in water positions (0.9 Å) and rather low conservation ratio (0.67). This cluster partially overlaps with three other large clusters that also have relatively low conservation ratios. This indicates that the water positions in this region of the MHC molecule are rather variable and that our clustering procedure had some difficulty in clustering them.
Figure 2 shows the relationship between the average normalized B-factors of the water positions in each cluster and two other parameters of the clusters, the conservation ratio and the average distance between the positions. The normalized B-factor can be taken as an indicator of the accuracy with which the water positions in the cluster have been determined in the crystallographic experiment (Buerger, 1960). It was computed as
 |
(B) are the average and the standard deviation of the distribution of the B-factor values, respectively, in each protein structure (Parthasarathy and Murthy, 1997; Carugo and Argos, 1998; Carugo, 1999).
|
We see that clusters with high conservation ratios also display low average normalized B-factor values (Figure 2a
), indicating that the corresponding water positions are more accurately defined. In particular, clusters with conservation ratios of 0.8 have normalized B-factors <–1, indicating that they have B-factor values below average for their structure. Figure 2b shows that the normalized B-factor of clusters also correlates with their average dispersion or cluster size, with the better correlation occurring for clusters having smaller normalized B-factors.
These results taken together indicate that the more accurately a water position is defined in a given protein crystal structure, the more likely it is to be identified in a similar position in a different crystal structure of the same or closely related polypeptide. This agrees with findings made previously in studies of conserved water positions of other protein systems (Loris et al., 1999).
Detailed analysis of specific water positions
Of the total of 67 water positions conserved in more than half of the structures (Figure 1), three of the most conserved positions (labeled 1, 2 and 3 in Table II) warrant detailed analysis. Their location relative to the MHC backbone is depicted in Figure 3, which displays the MHC molecule with the TCR binding side facing the viewer.
|
The first water molecule is located on the side of the ß-sheet lining the floor of the peptide binding groove and forms hydrogen bonds with nearby main chain atoms of the sheet. The second water molecule is located in the peptide-binding groove, near the N-terminus of the bound peptide. This water molecule makes hydrogen bonds with residues lining pocket B in the peptide-binding groove as shown in Figure 3
. The third water molecule is located near the broken -helix in the 2-domain and seems to play a role in determining the position of the N-terminus of -helix in this domain. In the following we analyze in detail the environment of these three water molecules and discuss their putative role in the context of the MHC–peptide complex.
The most highly conserved water position and its putative structural role.
The most highly conserved water position, present in all 12 MHC structures, is represented by cluster No. 1. Its hydrogen bonding interactions with the main chain atoms in the structure of HLA-B3501 (1A1N) (see Figure 4) are as follows: O water–O Ile94 (2.94 Å), O water–OE1 Gln96 (2.89 Å) and O water–N Gly120 (2.77 Å).
|
Of these, the interaction with the side chain of Gln96 warrants a special comment. The OE1 and NE2 of this Gln96 interact with NE2 His31 and O Trp60 in ß2-microglobulin in all 12 structures. As verified using a Blast search (Altschul et al., 1997
), this residue is strictly conserved in all the MHC class-I homologues. In addition, we see that its side chain also adopts a virtually identical conformation in all 12 structures, as illustrated in Figure 5, and that the O water–OE1 Gln96 H-bond is also completely conserved. Since the same water molecule is furthermore hydrogen bonded to oxygen atoms of the ß-sheet backbone (see below), this molecule can be viewed as playing a role in constraining the Gln side chain to lean against the face of the sheet.
|
Interestingly, the same Gln96 residue also makes two H-bonds with the imidazole group of His31 and the carbonyl oxygen of Trp60 of the ß2-microglobulin molecule, which forms a complex with the MHC moiety in the 1A1N RCSB-PDB entry. Since the association between ß2-microglobulin and the MHC class-I molecules is believed to be essential for peptide antigen presentation by these molecules (Boyd et al., 1992
; Wang et al., 1994; Balendiran et al., 1997), this conserved water molecule probably also plays a role in this association and thereby indirectly influences peptide binding.
A third interesting observation can be made concerning the interactions of water in this conserved position with the protein backbone. Inspection of the backbone conformation in this region reveals that the polypeptide segment spanning residues 118–121 adopts a twisted ß-turn conformation (Figures 4 and 5), with the following backbone torsion angles: (
118, 
118) = (–127.8°, 122.3°), (119, 119) = (50.7°, 35.8°), (120, 120) = (80.5°, 11.9°) and (121, 121) = (–126.4°, 140.9°). This segment is located near the C- and N-termini of -helices in 1- and 2-domains, respectively. The high degree of conservation in our structure set of both the water position and the twisted ß-turn conformation of the backbone segment with which it interacts suggests that the water molecule plays a role in stabilizing this conformation.
To verify this hypothesis a database comprising 510 451 pentapeptide fragments from protein structures in the PDB with <90% sequence identity (Ogata and Umeyama, 1998) was searched for fragments with a similar twisted ß-turn structure. This search was performed by superimposing all the pentapeptides from the database on to the 118–121 segment of the 1A1N entry and selecting those which superimposed with an r.m.s.d. 
1.0 Å. Water molecules within a 5.0 Å distance from the selected fragments in the parent structures were extracted in turn and their coordinates were transformed into the common reference frame. From those, water molecules within 1.5 Å of the conserved water position in MHC were retained and their interactions examined.
This search produced 59 fragments from different protein structures including six MHC class-I molecules; 43 of those, including four fragments from MHC molecules (located elsewhere than residues 118–122) have a water molecule near the main chain of residue n + 3 of the ß-turn. These are shown superimposed on to the 118–122 fragment of MHC together with the corresponding water molecules in Figure 6. This shows that twisted ß-turn structures in very different protein structures tend to have a water molecule interacting with the backbone of residue n + 3, suggesting that this water molecule plays an important role in maintaining this conformation. Its conservation in the MHC class-I molecules observed here can hence also be attributed to a structural role of a rather general nature.
|
Conserved water molecules near the bound peptide. Figures 7 and 8
illustrate the surroundings and interactions of the water molecule corresponding to the second most conserved water position (cluster No. 2), occurring in all structures except in the ‘non-classical’ class I MHC molecule (PDB code, 1MHC). Its hydrogen-bonding network in the HLA-B3501 structure (1A1N) comprises the following H-bonds: O water–OH Tyr7 (2.64 Å), O water–OH Tyr59 (2.86 Å) and O water–OD1 Asn63 (2.69 Å). This water position is located at the vicinity of the N-terminus of the bound peptide (Figures 7 and 8), but too far from it (5.05 Å) to form a hydrogen bond. However, the Tyr7 side chain with which this water interacts does form an H-bond with the peptide N-terminus in the majority of the MHC class-I structures analyzed in this work, albeit not in the 1AlN entry, used as reference in our detailed analysis of the atomic contacts. This water position therefore seems to correspond to the water molecule previously identified in the HLA-B27 structure by Madden et al. (Madden et al., 1992). Its absence in the non-classical 1MHC entry is most probably due to the fact that this protein is specific for N-formylated peptides only.
|
|
According to Zhang et al. (Zhang et al., 1998
), the most common amino acids at position 63 in human and murine MHC class-I sequences are Glu (60.7%), Asn (26.2%), Gln (7.7%) and Ile (5.4%) and this residue interacts with the P1 and P2 sites of the bound peptide. Of the 12 MHC molecules analyzed here, five (1I4F, 1HSA, 1QLF, 1QO3 and 1KBG) have Glu at position 63, five (1TMC, 1AGD, 1A1N, 1E27 and 1A1M) have Asn, one (1ED3) has Gln and one (1MHC) has Leu. Inspection of the Glu-containing structures reveals that this side chain adopts a very similar conformation in these structures, which fosters H-bond formation with the conserved water molecule. The side chains of Asn at position 63 in the other four structures and of Gln in 1ED3 also display similar orientations and form an H-bond with the conserved water molecule. On the other hand, the presence of Ile at position 63 in 1MHC is consistent with the absence of this particular water molecule in this crystal structure. Indeed, the non-polar Ile side chain cannot undergo polar interactions with water, but has been observed to form hydrophobic interactions with the Pro side chain of the peptide at the P2 site (Zhang et al., 1998). These features of the H2-M3 murine MHC protein are most probably related to its well-defined role of binding N-formylated peptide.
These observations and considerations suggest that when position 63 is occupied by a polar residue Glu, Gln or Asn, a ‘structural’ water molecule plays a part in modulating the interactions of the MHC molecules with the P1 and P2 sites of the peptide. It is interesting to see in this regard that, unlike for position 63, the side chains at positions 7 and 59 (both tyrosines) with which the conserved water molecule also forms H-bonds are completely conserved in type and conformation in our sample of MHC molecules, as illustrated in Figure 8.
Conserved water position at the first -helix kink in the 2-domain.
The third most conserved water position, corresponding to cluster No. 3, occurs in 10 out of the 12 analyzed structures (Table II). Figure 9 illustrates its environment in the HLA-B3501 (1A1N) structure. The H-bonds made by this water molecule are O water–NE1 Trp133 (3.05 Å), O water–O Trp147 (3.04 Å) and O water–N Asn153 (2.93 Å).
|
This conserved water molecule bridges two
-helical segments (residues 139–149 and 153–174), separated by three residues, which form a kink in the helix. This broken -helical motif seems to be specific to MHC molecules, since similar conformations could not be found in other proteins of the PDB using the fragment search procedure described above. Comparisons of the conformations of the different peptides forming complexes with the same MHC molecule have suggested that differences in backbone conformation between the MHC binding sites are primarily located in this kink region (Fremont et al., 1992; Madden et al., 1993; Ciatto et al., 2001). The backbone of Trp147, which is located near residues 97 and 116, composing pocket F (Zhang et al., 1998), forms a hydrogen bond with the conserved water molecule. This Trp residue is highly conserved among the all MHC class-I molecules. Moreover, the NE1 of Trp147 forms an H-bond with the backbone oxygen of P
-1 in the bound peptide, which is also well conserved (Madden, 1995; Ciatto et al., 2001).
To illustrate this point, Figure 10 displays the backbones of residues 152 and 153, the full Trp147 residue and the positions of conserved water molecules in all the MHC structures of our sample, after their backbones have been superimposed. The main chain atoms of P-1 in the peptide moieties are also displayed. We see that the displacement of the backbone oxygen of Trp147 is significantly larger than those of main chain atoms of residues 152 and 153 and reaches 2.61 Å. On the other hand, this displacement seems to follow closely that of main chain atoms of P-1 in the peptide. Furthermore, the hydrogen bond made by the backbone oxygen of Trp 147 to the conserved water molecule is maintained in all the MHC structures.
|
There therefore seems to be a correlation between the positions of the conserved water molecule and of the backbone oxygen of Trp147 and peptide binding. In bridging the two helical segments, the water molecules corresponding to cluster No. 3 may thus indirectly influence peptide binding, by influencing the position of the Trp147 side chain and thereby presumably regulating the size of the F peptide binding pocket, which is complementary to the amino acid residue at the P
-1 site. |
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
This study involved an analysis of conserved water position in a sample of 12 high-resolution MHC class-I molecules. Using an automatic clustering procedure, a total of 67 clusters of water molecules occurring in equivalent spatial positions in more than half of the analyzed structures were identified. These clusters were interpreted as representing conserved water positions in the corresponding MHC molecules. The normalized B-factor of the water molecules in each cluster was shown to correlate inversely with the extent to which these molecules are conserved in our sample of MHC structures and to correlate directly with the spread in their spatial positions in the different structures. The most highly conserved water positions in structures of our sample are therefore also those that were defined with the highest accuracy in the X-ray diffraction experiments. The number of conserved water positions in individual MHC molecules was furthermore found to be unrelated to the species from which the molecules originate. Structural data on other MHC–peptide complexes from different alleles and species, when they become available, will therefore most likely not alter the main conclusions of the present analysis.
The structural and functional roles of the three most conserved water positions have been discussed in detail. One appears to play an important structural role. Not only does it appear to stabilize a twisted ß-turn structure that occurs in all the MHC structures of our sample, but similar twisted ß-turn motifs, containing an equivalent water position, are also identified in many other protein structures of the PDB. This water molecule therefore seems to be a universal structural determinant of this specific and recurrent twisted ß-turn motif.
The two other water molecules are suggested to play a role in modulating, directly or indirectly, the positions and orientation of residues lining pockets in the binding groove of MHC class-I molecules, which are important for peptide recognition. One water molecule forms a hydrogen bond with an MHC residue influencing the positions of (MHC or peptide) residues in pocket B. The other water molecule bridges two segments of a kinked -helix in the 2 domain, whose relative orientation and position are likely to modulate the size of pocket F, thereby presumably indirectly influencing peptide binding.
The information provided here on the conserved water positions and their structural and functional roles should be very valuable for the prediction and modeling of MHC–peptide interactions, a topic with many useful practical applications. Our findings on the commonly occurring twisted ß-turn with its associated water molecule in proteins should have useful general applications in protein structure predictions.
|
Notes |
|---|
3 To whom correspondence should be addressed. E-mail: shosh{at}ucmb.ulb.ac.be
|
Acknowledgments |
|---|
This work was carried out as part of a project entitled ‘A multi-disciplinary approach to the development of epitope-based vaccines,’ funded by the European Communities, grant No. BIO4CT980294. We are grateful to all the members of this project for useful discussions.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConclusionReferences |
|---|
Altschul,S.F., Madden,T.L., Schaffer,A.A., Zhang,J., Zhang,Z., Miller,W. and Lipman,D.J. (1997) Nucleic Acids Res., 25, 3389–3402.
Balendiran,G.K., Solheim,J.C., Young,A.C., Hansen,T.H., Nathenson,S.G. and Sacchettini,J.C. (1997) Proc. Natl Acad. Sci, USA, 94, 6880–6885.
Berman,H.M., Westbrook,J., Feng,Z., Gilliland,G., Bhat,T.N., Weissig,H., Shindyalov,I.N. and Bourne,P.E. (2000) Nucleic Acids Res., 28, 235–242.
Bjorkman,P.J., Saper,M.A., Samraoui,B., Bennett,W.S., Strominger,J.L. and Wiley,D.C. (1987a) Nature, 329, 512–518.[CrossRef][Medline]
Bjorkman,P.J., Saper,M.A., Samraoui,B., Bennett,W.S., Strominger,J.L. and Wiley,D.C. (1987b) Nature, 329, 506–512.[CrossRef][Medline]
Boyd,L.F., Kozlowski,S. and Margulies,D.H. (1992) Proc. Natl Acad. Sci. USA, 89, 2242–2246.
Buerger,M.J. (1960) Crystal Structure Analysis. Wiley, New York.
Carugo,O. (1999) Protein Eng., 12, 1021–1024.
Carugo,O. and Argos,P. (1998) Proteins, 31, 201–213.[CrossRef][Web of Science][Medline]
Ciatto,C., Tissot,A.C., Tschopp,M., Capitani,G., Pecorari,F., Pluckthun,A. and Grutter,M.G. (2001) J. Mol. Biol., 312, 1059–1071.[CrossRef][Web of Science][Medline]
Collins,E.J., Garboczi,D.N., Karpusas,M.N. and Wiley,D.C. (1995) Proc. Natl Acad. Sci. USA, 92, 1218–1221.
Fremont,D.H., Matsumura,M., Stura,E.A., Peterson,P.A. and Wilson,I.A. (1992) Science, 257, 919–927.
Fremont,D.H., Stura,E.A., Matsumura,M., Peterson,P.A. and Wilson,I.A. (1995) Proc. Natl Acad. Sci. USA, 92, 2479–2483.
Fremont,D.H., Hendrickson,W.A., Marrack,P. and Kappler,J. (1996) Science, 272, 1001–1004.[Abstract]
Fremont,D.H., Monnaie,D., Nelson,C.A., Hendrickson,W.A. and Unanue,E.R. (1998) Immunity, 8, 305–317.[CrossRef][Web of Science][Medline]
Garboczi,D.N., Ghosh,P., Utz,U., Fan,Q.R., Biddison,W.E. and Wiley,D.C. (1996) Nature, 384, 134–141.[CrossRef][Medline]
Ghosh,P., Amaya,M., Mellins,E. and Wiley,D.C. (1995) Nature, 378, 457–462.[CrossRef][Medline]
Glithero,A. et al. (1999) Immunity, 10, 63–74.[CrossRef][Web of Science][Medline]
Hillig,R.C., Coulie,P.G., Stroobant,V., Saenger,W., Ziegler,A. and Hulsmeyer,M. (2001) J. Mol. Biol., 310, 1167–1176.[CrossRef][Web of Science][Medline]
Kendall,M.G. (1975) Multivariate Analysis. Charles Griffin, London.
Kryger,G., et al. (2000) Acta Crystallogr. D Biol. Crystallogr. 56, 1385–1394.
Loris,R., Langhorst,U., De Vos,S., Decanniere,K., Bouckaert,J., Maes,D., Transue,T.R. and Steyaert,J. (1999) Proteins, 36, 117–134.[CrossRef][Web of Science][Medline]
Madden,D.R. (1995) Annu. Rev. Immunol., 13, 587–622.[CrossRef][Web of Science][Medline]
Madden,D.R., Gorga,J.C., Strominger,J.L. and Wiley,D.C. (1992) Cell, 70, 1035–1048.[CrossRef][Web of Science][Medline]
Madden,D.R., Garboczi,D.N. and Wiley,D.C. (1993) Cell, 75, 693–708.[CrossRef][Web of Science][Medline]
Maenaka,K., Maenaka,T., Tomiyama,H., Takiguchi,M., Stuart,D.I. and Jones,E.Y. (2000) J. Immunol., 165, 3260–3267.
Matsumura,M., Fremont,D.H., Peterson,P.A. and Wilson,I.A. (1992) Science, 257, 927–934.
Meng,W.S., von Grafenstein,H. and Haworth,I.S. (1997) Int. Immunol., 9, 1339–1346.
Meng,W.S., von Grafenstein,H. and Haworth,I.S. (2000) Int. Immunol., 12, 949–957.
Murthy,V.L. and Stern,L.J. (1997) Structure, 5, 1385–1396.[Medline]
Ogata,K. and Umeyama,H. (1998) Proteins, 31, 355–369.[CrossRef][Web of Science][Medline]
Parthasarathy,S. and Murthy,M.R. (1997) Protein Sci., 6, 2561–2567.[Web of Science][Medline]
Rashin,A.A., Iofin,M. and Honig,B. (1986) Biochemistry, 25, 3619–3625.[CrossRef][Medline]
Reid,S.W. et al. (1996) J. Exp. Med., 184, 2279–2286.
Rothbard,J.B. and Gefter,M.L. (1991) Annu. Rev. Immunol., 9, 527–565.[CrossRef][Web of Science][Medline]
Saper,M.A., Bjorkman,P.J. and Wiley,D.C. (1991) J. Mol. Biol., 219, 277–319.[CrossRef][Web of Science][Medline]
Scott,C.A., Peterson,P.A., Teyton,L. and Wilson,I.A. (1998) Immunity, 8, 319–329.[CrossRef][Web of Science][Medline]
Sewell,A.K., Price,D.A., Oxenius,A., Kelleher,A.D. and Phillips,R.E. (2000) Stem Cells, 18, 230–244.
Smith,K.J., Reid,S.W., Harlos,K., McMichael,A.J., Stuart,D.I., Bell,J.I. and Jones,E.Y. (1996a) Immunity, 4, 215–228.[CrossRef][Web of Science][Medline]
Smith,K.J., Reid,S.W., Stuart,D.I., McMichael,A.J., Jones,E.Y. and Bell,J.I. (1996b) Immunity, 4, 203–213.[CrossRef][Web of Science][Medline]
Sneath,P.H.A. and Sokal,R.R. (1973) Numerical Taxonomy. Freeman, San Francisco.
Speir,J.A., Abdel-Motal,U.M., Jondal,M. and Wilson,I.A. (1999) Immunity, 10, 51–61.[CrossRef][Web of Science][Medline]
Speir,J.A., Stevens,J., Joly,E., Butcher,G.W. and Wilson,I.A. (2001) Immunity, 14, 81–92.[CrossRef][Web of Science][Medline]
Sreenivasan,U. and Axelsen,P.H. (1992) Biochemistry, 31, 12785–12791.[CrossRef][Medline]
Stern,L.J., Brown,J.H., Jardetzky,T.S., Gorga,J.C., Urban,R.G., Strominger,J.L. and Wiley,D.C. (1994) Nature, 368, 215–221.[CrossRef][Medline]
Tormo,J., Natarajan,K., Margulies,D.H. and Mariuzza,R.A. (1999) Nature, 402, 623–631.[CrossRef][Medline]
Wang,C.R., Castano,A.R., Peterson,P.A., Slaughter,C., Lindahl,K.F. and Deisenhofer,J. (1995) Cell, 82, 655–664.[CrossRef][Web of Science][Medline]
Wang,H., Capps,G.G., Robinson,B.E. and Zuniga,M.C. (1994) J. Biol. Chem., 269, 22276–22281.
Williams,M.A., Goodfellow,J.M. and Thornton,J.M. (1994) Protein Sci, 3, 1224–1235.[Web of Science][Medline]
Zhang,C., Anderson,A. and DeLisi,C. (1998) J. Mol. Biol., 281, 929–947.[CrossRef][Web of Science][Medline]
Received August 7, 2001; revised April 11, 2002; accepted April 18, 2002.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Bos and J. Pleiss Conserved Water Molecules Stabilize the {Omega}-Loop in Class A {beta}-Lactamases Antimicrob. Agents Chemother., March 1, 2008; 52(3): 1072 - 1079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Mustata and J. M. Briggs Cluster analysis of water molecules in alanine racemase and their putative structural role Protein Eng. Des. Sel., March 1, 2004; 17(3): 223 - 234. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||