Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

NMR Structure of Lipoprotein YxeF from Bacillus subtilis Reveals a Calycin Fold and Distant Homology with the Lipocalin Blc from Escherichia coli

  • Yibing Wu ,

    Contributed equally to this work with: Yibing Wu, Marco Punta

    Current address: Pharmaceutical Chemistry Department, University of California San Francisco, San Francisco, California, United States of America

    Affiliations Department of Chemistry, State University of New York at Buffalo, Buffalo, New York, United States of America, Northeast Structural Genomics Consortium

  • Marco Punta ,

    Contributed equally to this work with: Yibing Wu, Marco Punta

    Current address: Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, United Kingdom

    Affiliations Department of Computer Science and Institute for Advanced Study, Technical University of Munich, Munich, Germany, Northeast Structural Genomics Consortium

  • Rong Xiao,

    Affiliations Center of Advanced Biotechnology and Medicine, Department of Molecular Biology and Biochemistry, Robert Wood Johnson Medical School, The State University of New Jersey, Piscataway, New Jersey, United States of America, Northeast Structural Genomics Consortium

  • Thomas B. Acton,

    Affiliations Center of Advanced Biotechnology and Medicine, Department of Molecular Biology and Biochemistry, Robert Wood Johnson Medical School, The State University of New Jersey, Piscataway, New Jersey, United States of America, Northeast Structural Genomics Consortium

  • Bharathwaj Sathyamoorthy,

    Affiliation Department of Chemistry, State University of New York at Buffalo, Buffalo, New York, United States of America

  • Fabian Dey,

    Current address: Division of Molecular Biosciences, Imperial College London, London, United Kingdom

    Affiliations Howard Hughes Medical Institute, Department of Biochemistry and Molecular Biophysics, Center for Computational Biology and Bioinformatics, Columbia University, New York, New York, United States of America, Northeast Structural Genomics Consortium

  • Markus Fischer,

    Current address: Schrödinger, New York, New York, United States of America

    Affiliations Howard Hughes Medical Institute, Department of Biochemistry and Molecular Biophysics, Center for Computational Biology and Bioinformatics, Columbia University, New York, New York, United States of America, Northeast Structural Genomics Consortium

  • Arne Skerra,

    Affiliation Munich Center for Integrated Protein Science, CIPS-M, and Lehrstuhl für Biologische Chemie, Technische Universität München, Freising-Weihenstephan, Germany

  • Burkhard Rost,

    Affiliations Department of Computer Science and Institute for Advanced Study, Technical University of Munich, Munich, Germany, Northeast Structural Genomics Consortium

  • Gaetano T. Montelione,

    Affiliations Center of Advanced Biotechnology and Medicine, Department of Molecular Biology and Biochemistry, Robert Wood Johnson Medical School, The State University of New Jersey, Piscataway, New Jersey, United States of America, Northeast Structural Genomics Consortium

  • Thomas Szyperski

    szypersk@buffalo.edu

    Affiliations Department of Chemistry, State University of New York at Buffalo, Buffalo, New York, United States of America, Northeast Structural Genomics Consortium

NMR Structure of Lipoprotein YxeF from Bacillus subtilis Reveals a Calycin Fold and Distant Homology with the Lipocalin Blc from Escherichia coli

  • Yibing Wu, 
  • Marco Punta, 
  • Rong Xiao, 
  • Thomas B. Acton, 
  • Bharathwaj Sathyamoorthy, 
  • Fabian Dey, 
  • Markus Fischer, 
  • Arne Skerra, 
  • Burkhard Rost, 
  • Gaetano T. Montelione
PLOS
x

Abstract

The soluble monomeric domain of lipoprotein YxeF from the Gram positive bacterium B. subtilis was selected by the Northeast Structural Genomics Consortium (NESG) as a target of a biomedical theme project focusing on the structure determination of the soluble domains of bacterial lipoproteins. The solution NMR structure of YxeF reveals a calycin fold and distant homology with the lipocalin Blc from the Gram-negative bacterium E.coli. In particular, the characteristic β-barrel, which is open to the solvent at one end, is extremely well conserved in YxeF with respect to Blc. The identification of YxeF as the first lipocalin homologue occurring in a Gram-positive bacterium suggests that lipocalins emerged before the evolutionary divergence of Gram positive and Gram negative bacteria. Since YxeF is devoid of the α-helix that packs in all lipocalins with known structure against the β-barrel to form a second hydrophobic core, we propose to introduce a new lipocalin sub-family named ‘slim lipocalins’, with YxeF and the other members of Pfam family PF11631 to which YxeF belongs constituting the first representatives. The results presented here exemplify the impact of structural genomics to enhance our understanding of biology and to generate new biological hypotheses.

Introduction

The lipoprotein YxeF from Bacillus subtilis was selected by the Northeast Structural Genomics Consortium (NESG; http://www.nesg.org) as a target (gi|85674274, SwissProt/TrEMBL ID YXEF_BACSU, access number P54945, NESG target ID SR500A) of a biomedical theme project focusing on the structure determination of the soluble domains of bacterial lipoproteins [1], [2]. YxeF exhibits no significant sequence similarity with any protein with known three-dimensional structure and is one of only eight members forming Pfam [3] family PF11631 for which no functional annotation is available (Pfam 26.0 release). All members of the family are from the genus Bacillus and present high sequence similarity to YxeF, with A7ZAF5 and E1UTS8 from B. amyloliquefaciens being the most distant homologues (61% sequence identity to YxeF over 129 residues).

Bacterial lipoproteins represent a class of secreted, membrane-anchored proteins that are conserved throughout bacteria and play critical roles in a wide range of biological processes, including bacterial pathogenesis and host immune response [1]. They contain a conserved N-terminal type II signal peptide also known as the ‘lipobox’, which is immediately followed by an invariant Cys residue. After cleavage of the signal peptide, the lipoprotein is anchored into the bacterial membrane via a diacylglycerol moiety forming a thioether linkage to the Cys side chain as well as a fatty acid moiety coupled to the N-terminus.

In general, lipobox sequences exhibit the consensus sequence [Leu/Val/Ile]-[Ala/Ser/Thr/Val/Ile]-[Gly/Ala/Ser]-[Cys] (Val-Ser-Gly-Cys in YxeF). The residue following the Cys is important for localization of the lipoprotein [1]. In Gram-negative bacteria the lipoprotein is usually anchored in the inner membrane if Cys is followed by Asp, whereas otherwise it is anchored in the inner leaflet of the outer membrane [2]. Although Cys is followed by Gln in YxeF, it is supposedly anchored in the only lipid membrane of the Gram-positive Bacillus subtilis cell. Here we report the high-quality NMR solution structure of the soluble domain of protein YxeF comprising residues 19–144, along with a structural bioinformatics analysis to classify its structure and to gain new insights into its evolutionary origin.

Results and Discussion

NMR Structure of the Soluble Domain of Lipoprotein YxeF

A high-quality NMR structure of the soluble domain (comprising residues 19–144) of lipoprotein YxeF was obtained by multidimensional NMR spectroscopy [Figures 1A, 2; Table 1; PDB (protein data bank) accession number 2JOZ]. For simplicity, we henceforth refer to the structure of this domain as to the ‘YxeF structure’. YxeF contains nine β-strands (A to D and D’ to H) comprising residues 36–40, 52–57, 61–68, 72–75, 79–86, 91–97, 104–109, 114–119, and 122–129, respectively (Figure 2; strand assignment according to STRIDE [4]). Since strands D and D’ point into the same direction and are connected by a short coil region in a mostly extended conformation (comprising Pro 78), we will refer to this entire polypeptide segment as strand D (Figure 2). All β-strands are then arranged in anti-parallel fashion forming a +1 up-and-down β-barrel and are connected by seven loops L1 to L7. The β-barrel is closed on one side (loops L2, L4, L6), primarily by dense side chain packing of a number of hydrophobic and aromatic residues, including Phe 33, Tyr 34, Tyr 35, Trp 38 located immediately upstream of or on β-strand A, and additionally Tyr 81 and Leu 115. On the other side the β-barrel is open to the solvent (loops L1, L3, L5, L7) and lines a cavity with overall negative charge, predominantly due to the presence of Glu 40 on β-strand A, and Glu 64 and Glu 66 on β-strand C (Figure 3A,C).

thumbnail
Figure 1. 2D [15N,1H] HSQC spectra of lipoprotein YxeF.

(A) Spectrum recorded for the sample used for NMR structure determination at 750 MHz 1H resonance frequency. Resonance assignments are indicated using the one-letter amino acid code. Signals arising from side chains (Asn Hδ2/Nδ2, Gln Hε2/Nε2, Arg Hε/Nε and Trp Hε1/Nε1) are labeled with (*) and folded signals are designated with (†) next to the residue number. Signals arising from the His purification tag were not sequence specifically assigned. The spectral region indicated by dotted lines comprises most of the signals arising from the β-barrel (Figure 2) and is displayed for the spectra shown in (B). Those were recorded at different temperatures at 500 MHz 1H resonance frequency (see text).

https://doi.org/10.1371/journal.pone.0037404.g001

thumbnail
Figure 2. NMR structure of the soluble domain of lipoprotein YxeF.

Only residues 34–132 are shown because the terminal polypeptide segments are flexibly disordered in solution. The N-terminal residue Phe 34 and the C-terminal residue Ser 132 are labeled. Lines point at the Ω-type loop, which connects β-strands A and B and is poorly defined in the NMR structure. (A) Ribbon drawing of the first conformer of PDB entry 2JOZ with β-strands being depicted in cyan. (B) Superposition of the 20 conformers representing the NMR solution structure obtained after superposition of the backbone heavy atoms (N, Cα and C’) of the β-strands. (C) “Sausage” representation of backbone and best-defined side chains: a spline function was drawn through the mean positions of Cα atoms with the thickness being proportional to their mean global displacement in the 20 conformers after superposition as in B). The figure was generated using the program MOLMOL [59].

https://doi.org/10.1371/journal.pone.0037404.g002

thumbnail
Figure 3. Comparison of B. subtilis YxeF NMR structure and B. amyloliquefaciens A7ZAF5 homology model.

Surface electrostatic potential calculated for (A) the YxeF NMR structure (first conformer of ensemble deposited in the PDB) and (B) the homology model of A7ZAF5 by using the program GRASP [56] accessed through the protein function annotation server MarkUs [55]. The homology model was calculated using the SWISS-MODEL server in alignment mode [60], [61] and Verify3D [63], Procheck [64] and ProsaII [65] all atom z-scores (-1.12, −3.43 and −1.61, respectively) were obtained using the PSVS server [66] and are indicative of a good quality model. In (C) and (D), ribbon drawings are shown for the structures of YxeF and A7ZAF5 in the same orientation, that is, viewed on the open end of the β-barrels. The acidic residues giving rise to the negative potential inside the cavities are depicted in licorice representation and are labeled (black for YxeF, red for A7ZAF5). (E) Pfam multiple alignment of the sequences of all members of PF11631. Except for YxeF (P54945), the sequences are labeled with their UniProt [25] IDs (D4G3V0, E8VFY0, E0TYE6, D5MWC1, E3E109, A7ZAF5, E1UTS8). Amino acid background colors reflect average similarity inferred from the Blosum62 matrix, ranging from ‘most conserved’ (black) to ‘least conserved’ (white). YxeF and A7ZAF5 are highlighted in bold on the left and the region of the alignment used for building the comparative model of A7ZAF5 from the YxeF structure is enclosed by red boxes. The acidic residues labeled in (C) and (D) are marked with black (YxeF) and red (A7ZAF5) asterisks, respectively, above or below the alignment.

https://doi.org/10.1371/journal.pone.0037404.g003

Current Classification of YxeF Structure in the CATH, SCOP and Pfam Databases

Inspection of the YxeF structure (Figure 2) shows that it resembles β-barrel proteins belonging to the ‘calycin superfamily’ which includes lipocalins, fatty acid binding proteins, triabin, avidins/streptavidins and a class of metalloprotease inhibitors. All calycins contain a calyx-like β-barrel characterized by a +1 up-and-down topology (Figure 4), with triabin being the only exception due to a β-strand swap, and fatty acid-binding proteins featuring two additional β-strands in the barrel with respect to other calycins (i.e., 10-stranded instead of 8-stranded) [5], [6]. The β-barrels structurally characterizing calycins are open to the solvent on one side and often harbor a ligand-binding site [6], [7].

thumbnail
Figure 4. Schematic representation of secondary structure element topologies.

(A) YxeF, (B) lipocalins and (C) fatty acid-binding proteins. β-strands are represented by arrows, α-helices by rectangles, and 310-helices by ellipses. N- and C-termini are indicated as N and C respectively, and the ‘Ω-type’ loop L1 shared by YxeF and lipocalins is labeled.

https://doi.org/10.1371/journal.pone.0037404.g004

Accordingly, our YxeF structure (Figure 2) has been incorporated in the CATH (class architecture topology homologous superfamily) and SCOP (structurally classification of proteins) databases [8], [9]. In CATH, it is part of the Homology sub-level 2.40.128.20 within the ‘lipocalin’ Topology. This Homology sub-level incorporates lipocalins, fatty acid binding proteins and triabin, while the ‘lipocalin’ Topology includes all other calycins together with additional members such as some outer membrane proteins. In SCOP, YxeF is assigned to the ‘retinol binding protein-like’ family, containing all lipocalins of known structure. This family is found within the ‘lipocalin’ SCOP superfamily further including fatty acid binding proteins and triabin. Avidin/streptavidin and metalloprotease inhibitors are instead assigned to a different SCOP fold (i.e., ‘streptavidin-like’). Finally, in the Pfam sequence database lipocalins are grouped with fatty acid-binding proteins in several families within the ‘calycin superfamily’ clan [10], which additionally includes triabin. Avidins/streptavidins and metalloprotease inhibitors are not considered to be part of the ‘calycin superfamily’. These classifications are (i) based on both sequence and structure comparisons, (ii) rely, at least to some degree, on manual curation, and (iii) favor the hypothesis that an evolutionary link exists between lipocalins, fatty acid binding proteins and triabin. They leave, however, the tetrameric avidins/streptavidins and some metalloprotease inhibitors in limbo with respect to their relationship to the other proteins alluded to above.

SCOP further identifies lipocalins as a sub-group of more closely related proteins and places YxeF among them. Lipocalins are extracellular (sometimes membrane anchored) proteins known to generally transport and store small, largely hydrophobic compounds within a ligand pocket surrounded by four loops at the open end of the β-barrel [5], [11]. Despite sharing with lipocalins the same β-barrel topology YxeF lacks a C-terminal α-helix (Figure 4A,B) which, in all lipocalins with known structure, packs against one side of the β-barrel. This observation raises the question of whether and how YxeF is evolutionary related to lipocalins. One of the key challenges associated with classifying calycin−/lipocalin-like proteins is their typically very low (i.e., insignificant) sequence identity, so that quite often homology cannot be inferred from sequence alone [5], [6]. Furthermore, the manifold of known eight stranded β-barrels appears to form what has been named a structural ‘quasi-continuum’ [12]. This greatly impedes the identification of boundaries between divergent and convergent evolutionary links. In the following, we present a structural bioinformatics analysis aimed at resolving the YxeF structure classification and elucidating YxeF’s evolutionary origin.

YxeF Structure Belongs to the Calycin Superfamiliy

Calycins feature a conserved Gly-X-Trp/Arg signature motif (Figure 5), in which the Arg side chain is located on strand H, interacts with the Trp side chain located on strand A and also forms hydrogen bonds with the backbone carbonyl groups of some other N-terminal residues [5], [11], [13]. Consistently, this motif has been shown to be important for protein stability in the retinol-binding protein, a prototypic member of the lipocalin family [14], [15]. In YxeF, the motif is entirely conserved (Gly 36, Trp38 and Arg 128), although the conformation of the side chain of Arg 128 is rather poorly defined in the NMR structure (Figure 5B). Conservation of the calycin signature motif and of the β-barrel topology (Figure 4) renders straightforward the classification of YxeF as a ‘calycin’.

thumbnail
Figure 5. Comparison of YxeF NMR structure (PDB ID 2JOZ, coded in blue) and Blc X-ray crystal structure (PDB ID 3MBT, orange).

(A) Structure-based sequence alignment between YxeF and Blc obtained with the program DALI [16]. The three structurally conserved regions (SCR1-3) typically found in lipocalins (see text) are boxed (continuous line for SCR1, which appears to be conserved in YxeF; dashed line for SCR2 and SCR3). Conserved residues being part of the calycin signature motif resulting in an interaction between Gly 36-X-Trp 38 in SCR1 and Arg 128 in SCR3 (see text) are highlighted using red boxes. Residues being part of the second hydrophobic core of Blc [see also (D] are highlighted using cyan boxes. (B) Superposition of the Trp and Arg residues being part of the calycin Gly-X-Trp and Arg motif in Blc (licorice representation, orange) and YxeF (line representation, all NMR conformers, blue). The superposition is obtained after superposition of the X-ray structure of Blc with each conformer of the NMR solution structure of YxeF (residues 32–132). (C) Structural superposition generated by the program DALI viewed from the open end of the β-barrels (for YxeF residues 32–132 were considered). In Blc, box 1 identifies the C-terminally located α-helix and box 2 the C-terminal β-strand, which are packed against the outside of the β-barrel and thereby form a second hydrophobic core (see D). (D) Ribbon drawing of the Blc structure with licorice representation of hydrophobic residues (in cyan) located in the C-terminal α-helix and on the outside of the β-barrel forming a second hydrophobic core [see also (C)].

https://doi.org/10.1371/journal.pone.0037404.g005

DALI-based Search for Similar Structures

A search of the PDB using the program DALI [16] for proteins which are structurally similar to YxeF yielded more than 700 significant hits (152 when considering PDB90 or a set of PDB proteins redundancy reduced at 90% sequence identity). These hits span a quasi-continuum of Z-scores between 7.7 (top hit) and 2.1 (taken as a lower limit of significance). Among the structurally similar proteins are calycins from all groups alluded to above. Top hits include the cysteine protease inhibitor staphostatin B (Z-score  = 7.7), several 10-stranded fatty acid binding proteins (top Z-score  = 7.5), the C-terminal domain of a self-compartmentalizing protease Pab87 from Pyrococcus abyssi (Z-score  = 7.2) which has been claimed [17] to be the first domain with lipocalin-like architecture from Archaea, and several lipocalins (top Z-score 6.9). Staphostatin B is classified in SCOP as having a ‘Streptavidin-like’ fold but high structural similarity to lipocalins was recognized previously [18]. It should be noted, however, that in spite of the high Z-score returned by overall structure superposition, topology of strand A to C in staphostatin’s β-barrel is very different with respect to the one observed in both streptavidins and lipocalins. Other calycin structures result in lower DALI Z-scores (avidin’s Z-score is 5.5).

Overall structural similarity identified with DALI thus seems to clearly indicate evolutionary relatedness between YxeF, fatty-acid binding proteins and lipocalins. Two structural features shared by YxeF and lipocalins, however, strongly suggest that YxeF is closer to the latter than to fatty-acid binding proteins (Figure 4): (i) the 8 strands forming the β-barrel (versus 10 in fatty acid-binding proteins), and (ii) the presence of an ‘Ω-type loop’ connecting β-strands A and B (in fatty-acid binding proteins two short α-helices are inserted between strands A and B). Notably, the Ω-type loop that often acts as a flexible lid for the open end of the β-barrel in lipocalins appears to be disordered in YxeF (Figure 2). On the other hand, as mentioned above, YxeF lacks two structural features conserved in the lipocalins (Figure 4): an N-terminal 310-helix, which is sometimes replaced by a longer α-helix [19], [20], and a C-terminally located α-helix followed by an additional β-strand [21], [22]. With respect to the first difference, several aromatic residues located in YxeF in the polypeptide segment preceding β-strand A (i.e., Phe 33, Tyr 34, Tyr 35) play a very similar structural role as the forming the 310-helix in lipocalins, that is, they contribute to occlude the bottom of the β-barrel. Thus, solely the absence of the C-terminal α-helix and β-strand remain as a stark structural difference when comparing the structure of YxeF with known lipocalin structures.

YxeF and Lipocalin Blc from E. coli are Distant Homologues

To further refine our structural analysis, we compared in detail the structure of YxeF with that of lipoprotein Blc from E. coli (Figure 5), the only bacterial lipocalin for which an atomic-resolution structure is currently available (PDB ID: 3MBT) [23]. It has been suggested that lipocalins can be grouped into ‘kernel’ or ‘outlier’ lipocalins [11] depending on the presence or absence of three so-called ‘structurally conserved regions’ (SCRs) which correlate to some degree with sequence conservation (Figure 5A). The most conserved SCR1 comprises the N-terminal Gly-X-Trp segment of the calycin Gly-X-Trp/Arg signature motif located on β-strand A, while SCR3 (formed primarily by residues found on β-strand H) contains the Arg residue as part of the same motif. SCR2, instead, spans the region between the termini of β-strands F and G including the loop L6 connecting the two β-strands. Because the SCR2 and SCR3 sequence motifs are not conserved in Blc, it was initially classified as an ‘outlier lipocalin’ [24]. However, this sequence-based classification does not appear to be well justified when considering that the X-ray structure of Blc revealed a remarkably close structural similarity with ‘kernel’ lipocalins [23].

Like YxeF, Blc is monomeric and does not contain disulfide bridges, which are frequently found in other lipocalins. While the overall structural alignment of YxeF and Blc results in a highly significant, but not the highest DALI Z-score [Z-score  = 6.9; 2.7 Å r.m.s.d. (root mean square deviation) for superposition of the Cα atoms of 88 aligned residues exhibiting 17% sequence identity], the structural similarity of the two β-barrels (48 residues) is truly striking: they superimpose with a backbone r.m.s.d. of only 1.8 Å (Figure 5). Indeed, the structural alignment of only the β-barrels yields the highest Z-score for Blc among a selection of full structure alignment top hits (Table 2; note that in these comparisons we consider conformer 1 of the ensemble representing the YxeF solution structure 2JOZ). The corresponding values calculated for avidin, which contains a β-barrel of evidently different shape (Figure 6), are also provided in Table 2 for comparison. The only minor structural difference between the YxeF and Blc β-barrels, possibly reflecting different ligand specificities, relates to β-strand A at the base of the Ω-type loop. This strand is shorter in YxeF where it creates a small V-shaped aperture on the side of the β-barrel (Figure 6).

thumbnail
Table 2. Comparison of β-barrels occurring in selected structuresa yielding top DALI Z-scores after full structure alignment with YxeF.

https://doi.org/10.1371/journal.pone.0037404.t002

thumbnail
Figure 6. Comparison of β-barrels.

Ribbon drawings of β-barrels of avidin (PDB ID 1AVD, green) in (A) and, after rotation by 180°, in (D); bacterial lipocalin Blc from E. coli (PDB ID 3MBT, orange) in (B) and (E); YxeF in (C) and (F) (PDB ID 2JOZ, blue). For clarity, the disordered terminal polypeptide segments of YxeF, as well as the corresponding segments in avidin and Blc, are not shown. In (A)–(C), β-strands A and H are labeled, while in (D)–(F) β-strand D is indicated.

https://doi.org/10.1371/journal.pone.0037404.g006

As indicated above, the C-terminally located α-helix characteristic of lipocalins is not present in YxeF (Figure 4). In Blc, this α-helix packs against the outside of the β-barrel, primarily against β-strands G and H (Figure 5C). As a result, a second hydrophobic core is formed, adding to the one found in the lower, closed part of the β-barrel itself (Figure 5D). In YxeF, the corresponding C-terminally located polypeptide segment is highly polar and flexibly disordered in solution. The absence of hydrophobic residues in this segment and also on the exterior of β-strands G and H apparently prevents the formation of a second hydrophobic core. Since this core has been shown to be important for the stability of lipocalins, and Blc exhibits a comparably low melting temperature of ∼45°C even with this second core, the stability of the well-defined fold of YxeF evidenced by our high-quality structure (Figure 2) is somewhat unexpected. Intriguingly, although the onset of protein precipitation at higher temperature prevented us from accurately determining a heat denaturation ‘melting’ temperature, the inspection of 2D [15N,1H] HSQC (heteronuclear single quantum coherence) spectra recorded up to ∼50°C revealed that YxeF’s β-barrel is intact even at such elevated temperatures (Figure 1B).

Taken together, in spite of the lack of the C-terminally located α-helix, the strong structural similarity of the β-barrels of YxeF and Blc (Table 2), together with a remarkably similar relative spatial orientation of the Trp and Arg residues of the Gly-X-Trp/Arg signature motif (Figure 5B), reveals that YxeF and lipocalin Blc are distant homologues. This is consistent with the fact that both YxeF and Blc are secretory lipoproteins, a characteristic common to most predicted lipocalins in Gram-negative Bacteria [24].

Homology Model of Protein A7ZAF5: Insights into Putative Ligand Binding

The only proteins known to share significant sequence identity with YxeF are found in the Bacillus genus, e.g., D5MWC1 from B. subtilis strain ATCC 6633, E3E109 form B. atrophaeus strain 1942, and the somewhat more distant homolog A7ZAF5 from B. amyloliquefaciens (61% identity; member of PF11631; see Figure 3E for a multiple sequence alignment of all members). In D5MWC1 and E3E109, the three glutamate residues that confer a negative charge distribution to YxeF’s putative ligand binding site (Figure 3A,C) are conserved. Despite the high overall sequence similarity, these glutamates are absent in A7ZAF5. Interestingly, however, in a homology model calculated for the soluble domain of A7ZAF5 based on conformer 1 of the YxeF NMR structure (2JOZ), it appears that other negatively charged residues, i.e., Glu 52, Asp 78, Glu 121 (residue numbers as in the UniProt [25] sequence of A7ZAF5) create a similar surface charge distribution within the corresponding cavity of A7ZAF5 (Figure 3B,D,E; note that Glu 121 is replaced by Gly in E1UTS8, while additionally conserved in E3E109). Considering the robustness of homology modeling at >60% pairwise sequence identity and the absence of any gaps in the alignment, this finding suggests that these proteins may function by binding yet to be identified ligands that share similar electrostatic properties.

Evolutionary Origin of Lipocalins and Protein YxeF

Lipocalins are a functionally diverse group of proteins that usually bind small hydrophobic molecules. Bacterial lipocalins, in particular, have been proposed to be implicated in biogenesis of the outer membrane of Gram-negative bacteria based on three findings: (i) Blc from E. coli has been shown to bind fatty acids and (lyso)-phospholipids [26], suggesting its participation in lipid metabolism [26], (ii) no lipocalin had so far been identified in microorganisms that lack an outer lipid membrane, that is, Gram-positive bacteria or Archaea, and (iii) some integral outer membrane proteins exhibit significant structural similarity with lipocalins [21], [27].

Our finding that YxeF and Blc are distant homologues suggests to classify YxeF (and thus all currently known members of PF11631 exhibiting high sequence similarity; Figure 3E) as a hitherto unknown type of lipocalin constituting a new sub-family characterized by its distinct ‘β-barrel only’ architecture. We suggest to name this new sub-family 'slim lipocalins’. It is then of key importance for our understanding of the evolution of lipocalins that YxeF is from the Gram-positive bacterium B. subtilis. It has been suggested that lipocalins first emerged along with the evolution of the outer membrane in Gram-negative bacteria [21]. The identification of distant homology between Blc and YxeF now suggests instead that lipocalins may have emerged before the divergence of Gram-positive and Gram-negative bacteria [28]. Although the presence of YxeF in B. subtilis might also have resulted from horizontal gene transfer, the genomic island predictor Alien_hunter [29] applied to genomic regions in and around the yxeF gene in both B. subtilis and B. amyloliquefaciens did not provide any support for this hypothesis. Hence, based on the evidence collected so far, it is thus indeed tempting to speculate that an ancient lipocalin, like protein YxeF devoid of a second hydrophobic core, evolved into both YxeF-like proteins of Gram-positive bacteria and the lipocalins present nowadays in Gram-negative bacteria and Eukaryotes. Additional structural variability of the β-barrel, required to convey specificity for other physiological ligands, might have co-evolved with the formation of the second hydrophobic core and, possibly, the disulfide bridges present in many eukaryotic lipocalins in order to maintain, or even increase protein stability [30].

As indicated above, the C-terminal domain of a self-compartmentalizing protease Pab87 from Pyrococcus abyssi has been suggested to be the first domain with lipocalin-like architecture from Archaea [17]. Notably, the Pab87 structure (i) ranks high among the top DALI structural matches for YxeF (Z-score 7.2; r.m.s.d. 2.8 Å, 76 aligned residues, 5% sequence identity), (ii) likewise represents a slim ‘β-barrel only’ structure, and (iii) contains a calycin signature motif (Gly-X-Tyr/Lys). However, the DALI Z-score for comparison of the β-barrels (Table 2) is comparably low, and the domain does not exhibit the ‘Ω-type’ loop characteristic of lipocalins. Consistently, its function was linked to protein oligomerization (and thus compartmentalization of the active site of the full-length protein) and not to the binding of a ligand in a fashion typically observed for lipocalins. Intriguingly, however, comparison of the structures of YxeF and the C-terminal domain of Pab87 indicates that possibly a very ancient lipocalin may have existed, even before the divergence of Bacteria and Archaea occurred.

Prospects for Protein Design

The discovery of the ‘slim’ lipocalin YxeF devoid of the C-terminal α-helix (and of the entire second hydrophobic core) as well as disulfide bridges also provides a promising new scaffold for the design of lipocalins with novel ligand binding functions, so-called ‘anticalins’ [31]. The four structurally variable loops that form the entrance to the ligand pocket at the open end of a lipocalin’s β-barrel share functional similarity with the six hypervariable loops (CDRs) of antibodies. When compared with immunoglobulins, however, lipocalins are much smaller (∼160–180 residues), comprise only a single polypeptide chain, and can be produced at high yields in microbial host cells. Using targeted randomization of the structurally variable loop region in combination with phage display selection, anticalins with novel specificities have been engineered for the high affinity complexation of both low molecular weight compounds and protein antigens [7], [31], [32].

In many cases the lipocalin β-barrel is thermally rather stable and tolerates a wide range of amino acid substitutions at the ligand-binding site. Melting temperatures of natural lipocalins are often above 70°C [30] and can range beyond 95°C, for example, for human tear lipocalin. As mentioned above, Blc is actually a notable exception having a melting temperature of just ∼45°C. Attempts to engineer anticalins that lack the C-terminal α-helix were thus far not successful. Hence, the soluble domain of YxeF may turn out to be a promising target for the design of a novel line of minimal ‘β-barrel-only’, or ‘slim’ anticalins that can be used as reagents for bioanalytical purposes or separation tasks. It is evident that NMR and X-ray crystallographic studies [23], [33] will continue to be of key importance for this endeavor. Finally, since endogenous lipocalins are believed to play a role in antibiotic resistance and activation of immunity in Gram-negative bacteria, lipocalins of Gram-positive bacteria might turn out to be relevant biomedical targets themselves, e.g., for the development of new antibiotics [24].

Conclusions

The structure of the soluble domain of lipoprotein YxeF from the Gram-positive B. subtilis revealed an unexpected distant homology with lipocalin Blc from Gram-negative E. coli. Because YxeF is devoid of a second hydrophobic core typical for all lipocalins, we propose to introduce a new lipocalin sub-family named the ‘slim lipocalins’, with the members of Pfam family PF11631 being the first known representatives. The identification of YxeF as the first lipocalin homologue from a Gram-positive bacterium has far reaching consequences for our understanding of the evolution of this important class of proteins: lipocalins may have emerged well before the evolutionary divergence of Gram-positive and Gram-negative bacteria. Furthermore, we expect that the discovery of the ‘slim lipocalin’ YxeF will impact design of new anticalins with prescribed binding specificities. The results presented in this publication thus exemplify the role of structural genomics to generate new biological hypotheses and to support protein design efforts.

Materials and Methods

NMR Sample Preparation

The soluble domain of protein YxeF (excluding the ‘lipobox’ signal sequence) was cloned, expressed and purified following standard protocols developed by the NESG for production of uniformly U-13C,15N and 5%-13C, U-15N-labeled protein samples [34], [35]. Briefly, the truncated yxeF gene from Bacillus subtilis containing residues 19–144 was cloned into a pET21 (Novagen) derivative, yielding plasmid SR500A-21.2. The resulting construct contains eight nonnative residues at the C-terminus (LEHHHHHH) that facilitate protein purification. This expression vector, NESG SR500A-21.2, has been deposited in the PSI Materials Repository (http://psimr.asu.edu/). E. coli BL21 (DE3) pMGK cells, a codon enhanced strain, were transformed with SR500A-21.2 and cultured in MJ9 minimal medium containing (15NH4)2SO4 and U-13C-glucose or 5% U-13C-glucose/95% unlabeled glucose [36]. U-13C, 15N and 5%-13C, U-15N-labeled yxeF protein was purified using an AKTAxpress (GE Healthcare) based two-step protocol consisting of IMAC (HisTrap HP) and gel filtration (HiLoad 26/60 Superdex 75) chromatography. The final yield of purified U-13C, 15N and 5%-13C, U-15N protein YxeF (>98% homogenous by SDS-PAGE; 16.1 kDa by MALDI-TOF mass spectrometry) was ∼63 mg/L and ∼23 mg/L, respectively. In addition, U-15N, 5% biosynthetically directed fractionally 13C-labeled samples were generated to stereo-specifically assign Val and Leu methyl groups [37]. The final concentrations of U-13C, 15N and 5%-13C, U-15N labeled YxeF protein samples were about 1.1 mM and 1.2 mM, respectively, in a 90% H2O/10% D2O solution containing 100 mM NaCl, 5 mM CaCl2, 10 mM DTT, 0.02% NaN3, 50 µM 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS) and 20 mM 2-(N-morpholino)ethanesulfonic acid (MES) at pH = 6.5. An isotropic overall rotational correlation time of 8.6 ns was inferred from 15N spin relaxation times indicating that the protein yxeF is monomeric in solution. This conclusion was confirmed by analytic gel-filtration (Agilent Technologies) followed by a combination of static light scattering and refractive index (Wyatt Technology).

NMR Spectroscopy

NMR spectra were recorded at 25°C on Varian INOVA 600 or 750 spectrometers equipped with cryogenic probes. Five through-bond correlated G-matrix Fourier transform (GFT) NMR experiments complemented by three-dimensional (3D) HNNCO as described [38][41], were collected for backbone and side chain resonance assignment (total measurement time: 81 hours). Simultaneous 3D 15N/13Caliphatic/13Caromatic-resolved [1H,1H] NOESY (nuclear Overhauser enhancement spectroscopy; mixing time: 60 ms; measurement time: 26 hours) was acquired to derive 1H–1H distance constraints [40]. Two-dimensional (2D) constant-time [13C,1H]-HSQC spectra were recorded as was described for the 5% fractionally 13C-labeled samples to obtain stereo-specific assignments for isopropyl groups of Val and Leu [41]. In order to assess thermal stability of protein YxeF, a series of 2D [15N,1H]-HSQC spectra were recorded for a ∼150 µM protein solution between 25°C and about 50°C on a Varian INOVA 500 spectrometer equipped with a conventional probe (total measurement time: 30 hours; Figure 1B). All spectra were processed and analyzed with the programs NMRPIPE and XEASY, respectively [42], [43].

Sequence specific backbone (1HN, 15N, 1Hα, 13Cα) and 1Hβ/13Cβ resonance assignments were obtained by using (4,3)D HNNCαβCα/CαβCα(CO)NHN and (4,3)D HαβCαβ(CO)NHN along with the program AUTOASSIGN [44], and polypeptide backbone 13C’ resonances were assigned using 3D HNNCO. More peripheral side chain chemical shifts were assigned with aliphatic (4,3)D HCCH and 3D 15N/13Caliphatic/13Caromatic-resolved [1H,1H]-NOESY (for details of NESG NMR protocols, see http://www.nmr2.buffalo.edu/nesg.wiki). Overall, assignments were obtained for 99% of the backbone (excluding the N-terminal NH3+, the Pro 15N and the 13C’ preceding prolyl residues; Figure 1A) and 13Cβ, and for 98% of the side chain chemical shifts (excluding Lys NH3+, Arg NH2, OH, side chain 13C’ and aromatic 13Cγ) which are assignable with the set of NMR experiments provided above. Furthermore, 79% of Val and Leu isopropyl moieties and 20% of β-methylene groups with non-degenerate proton chemical shifts were stereo-specifically assigned (Table 1). Chemical shifts were deposited in the BioMagResBank (accession code: 15211) [45]. 1H–1H upper distance limit constraints for structure calculations were extracted from NOESY (Table 1). In addition, backbone dihedral angle constraints were derived from chemical shifts using the program TALOS for residues located in well-defined secondary structure elements [46]. The programs CYANA and AUTOSTRUCTURE were used in parallel to assign long-range NOEs [47][50]. The final structure calculations were performed using CYANA followed by explicit water bath refinement using the program CNS [51]. NMR structure quality was assessed with the Protein Structure Validiation Software Suite (PSVS) and evaluated by structural genomics consortia, and RPF [52], [53]. The coordinates were deposited in the RCSB Protein Data Bank (PDB) with accession code 2JOZ [54]. Amino acid numbers in the PDB coordinate file are those of the soluble domain only, numbered as residues 2–127 (with residue 1 being the Met start residue of the recombinant protein). The residues of the soluble domain correspond in UniProt sequence P54945 to residues 19–144, which is the numbering used throughout the paper.

Structural Bioinformatics

In silico studies of YxeF were primarily performed using the MarkUs server integrating a variety of computational tools [55], including the programs DALI and Skan for identification of structural similarities and calculation of structural alignments [16], [56], [57]. Moreover, the DALI pairwise alignment server was used to refine structural comparisons [58]. The programs MOLMOL [59] and STRIDE [4] were used to identify the location of regular secondary structure elements. A homology model was obtained for protein A7ZAF5, that is, one of the most distant known sequence homolog of YxeF, by submitting the YxeF-A7ZAF5 BLAST pairwise sequence alignment to the SWISS-MODEL server in alignment mode [60], [61]. Given the high sequence identity between template and target (61%) and the absence of any gaps, the comparative spatial localization of acidic residues inside the β-barrel appears to be robust. Alien_hunter genomic island predictions were obtained via the EnsemblBacteria website [29], [62].

Acknowledgments

We thank R. Shastry, C. Ciccosanti, H. Janjua, and G.V.T. Swapna for contributions in sample preparation, and Dr. Masayori Inouye, Rutgers University, for advice regarding the biology and biochemistry of bacterial lipoproteins. Support was also obtained from the University at Buffalo’s Center for Computational Research.

Author Contributions

Conceived and designed the experiments: YW RX TBA BS GTM TS. Performed the experiments: YW RX TBA BS. Analyzed the data: YW MP RX TBA BS FD MF AS BR GTM TS. Contributed reagents/materials/analysis tools: MP RX TBA FD MF BR GTM TS. Wrote the paper: YW MP AS BR GTM TS.

References

  1. 1. Babu MM, Priya ML, Selvan AT, Madera M, Gough J, et al. (2006) A database of bacterial lipoproteins (DOLOP) with functional assignments to predicted lipoproteins. J Bacteriol 188: 2761–2773.MM BabuML PriyaAT SelvanM. MaderaJ. Gough2006A database of bacterial lipoproteins (DOLOP) with functional assignments to predicted lipoproteins.J Bacteriol18827612773
  2. 2. Tokuda H, Matsuyama S (2004) Sorting of lipoproteins to the outer membrane in E. coli. Biochim Biophys Acta 1693: 5–13.H. TokudaS. Matsuyama2004Sorting of lipoproteins to the outer membrane in E. coli.Biochim Biophys Acta1693513
  3. 3. Punta M, Coggill PC, Eberhardt RY, Mistry J, Tate J, et al. (2012) The Pfam protein families database. Nucleic Acids Res 40: D290–301.M. PuntaPC CoggillRY EberhardtJ. MistryJ. Tate2012The Pfam protein families database.Nucleic Acids Res40D290301
  4. 4. Heinig M, Frishman D (2004) STRIDE: a web server for secondary structure assignment from known atomic coordinates of proteins. Nucleic Acids Res 32: W500–502.M. HeinigD. Frishman2004STRIDE: a web server for secondary structure assignment from known atomic coordinates of proteins.Nucleic Acids Res32W500502
  5. 5. Flower DR, North AC, Sansom CE (2000) The lipocalin protein family: structural and sequence overview. Biochim Biophys Acta 1482: 9–24.DR FlowerAC NorthCE Sansom2000The lipocalin protein family: structural and sequence overview.Biochim Biophys Acta1482924
  6. 6. Grzyb J, Latowski D, Strzalka K (2006) Lipocalins - a family portrait. J Plant Physiol 163: 895–915.J. GrzybD. LatowskiK. Strzalka2006Lipocalins - a family portrait.J Plant Physiol163895915
  7. 7. Beste G, Schmidt FS, Stibora T, Skerra A (1999) Small antibody-like proteins with prescribed ligand specificities derived from the lipocalin fold. Proc Natl Acad Sci U S A 96: 1898–1903.G. BesteFS SchmidtT. StiboraA. Skerra1999Small antibody-like proteins with prescribed ligand specificities derived from the lipocalin fold.Proc Natl Acad Sci U S A9618981903
  8. 8. Greene LH, Lewis TE, Addou S, Cuff A, Dallman T, et al. (2007) The CATH domain structure database: new protocols and classification levels give a more comprehensive resource for exploring evolution. Nucleic Acids Res 35: D291–297.LH GreeneTE LewisS. AddouA. CuffT. Dallman2007The CATH domain structure database: new protocols and classification levels give a more comprehensive resource for exploring evolution.Nucleic Acids Res35D291297
  9. 9. Andreeva A, Howorth D, Chandonia JM, Brenner SE, Hubbard TJ, et al. (2008) Data growth and its impact on the SCOP database: new developments. Nucleic Acids Res 36: D419–425.A. AndreevaD. HoworthJM ChandoniaSE BrennerTJ Hubbard2008Data growth and its impact on the SCOP database: new developments.Nucleic Acids Res36D419425
  10. 10. Finn RD, Mistry J, Tate J, Coggill P, Heger A, et al. (2010) The Pfam protein families database. Nucleic Acids Res 38: D211–222.RD FinnJ. MistryJ. TateP. CoggillA. Heger2010The Pfam protein families database.Nucleic Acids Res38D211222
  11. 11. Flower DR (1996) The lipocalin protein family: structure and function. Biochem J 318 (Pt 1): 1–14.DR Flower1996The lipocalin protein family: structure and function.Biochem J 318 (Pt1)114
  12. 12. Sadreyev RI, Kim BH, Grishin NV (2009) Discrete-continuous duality of protein structure space. Curr Opin Struct Biol 19: 321–328.RI SadreyevBH KimNV Grishin2009Discrete-continuous duality of protein structure space.Curr Opin Struct Biol19321328
  13. 13. Flower DR, North AC, Attwood TK (1993) Structure and sequence relationships in the lipocalins and related proteins. Protein Sci 2: 753–761.DR FlowerAC NorthTK Attwood1993Structure and sequence relationships in the lipocalins and related proteins.Protein Sci2753761
  14. 14. Greene LH, Chrysina ED, Irons LI, Papageorgiou AC, Acharya KR, et al. (2001) Role of conserved residues in structure and stability: tryptophans of human serum retinol-binding protein, a model for the lipocalin superfamily. Protein Sci 10: 2301–2316.LH GreeneED ChrysinaLI IronsAC PapageorgiouKR Acharya2001Role of conserved residues in structure and stability: tryptophans of human serum retinol-binding protein, a model for the lipocalin superfamily.Protein Sci1023012316
  15. 15. Katakura Y, Totsuka M, Ametani A, Kaminogawa S (1994) Tryptophan-19 of beta-lactoglobulin, the only residue completely conserved in the lipocalin superfamily, is not essential for binding retinol, but relevant to stabilizing bound retinol and maintaining its structure. Biochim Biophys Acta 1207: 58–67.Y. KatakuraM. TotsukaA. AmetaniS. Kaminogawa1994Tryptophan-19 of beta-lactoglobulin, the only residue completely conserved in the lipocalin superfamily, is not essential for binding retinol, but relevant to stabilizing bound retinol and maintaining its structure.Biochim Biophys Acta12075867
  16. 16. Holm L, Kaariainen S, Rosenstrom P, Schenkel A (2008) Searching protein structure databases with DaliLite v.3. Bioinformatics 24: 2780–2781.L. HolmS. KaariainenP. RosenstromA. Schenkel2008Searching protein structure databases with DaliLite v.3.Bioinformatics2427802781
  17. 17. Delfosse V, Girard E, Birck C, Delmarcelle M, Delarue M, et al. (2009) Structure of the archaeal pab87 peptidase reveals a novel self-compartmentalizing protease family. PLoS One 4: e4712.V. DelfosseE. GirardC. BirckM. DelmarcelleM. Delarue2009Structure of the archaeal pab87 peptidase reveals a novel self-compartmentalizing protease family.PLoS One4e4712
  18. 18. Rzychon M, Filipek R, Sabat A, Kosowska K, Dubin A, et al. (2003) Staphostatins resemble lipocalins, not cystatins in fold. Protein Sci 12: 2252–2256.M. RzychonR. FilipekA. SabatK. KosowskaA. Dubin2003Staphostatins resemble lipocalins, not cystatins in fold.Protein Sci1222522256
  19. 19. Mans BJ, Ribeiro JM, Andersen JF (2008) Structure, function, and evolution of biogenic amine-binding proteins in soft ticks. J Biol Chem 283: 18721–18733.BJ MansJM RibeiroJF Andersen2008Structure, function, and evolution of biogenic amine-binding proteins in soft ticks.J Biol Chem2831872118733
  20. 20. Paesen GC, Adams PL, Harlos K, Nuttall PA, Stuart DI (1999) Tick histamine-binding proteins: isolation, cloning, and three-dimensional structure. Mol Cell 3: 661–671.GC PaesenPL AdamsK. HarlosPA NuttallDI Stuart1999Tick histamine-binding proteins: isolation, cloning, and three-dimensional structure.Mol Cell3661671
  21. 21. Bishop R, Cambillau C, Privé G, Hsi D, Tillo D, et al. (2000) Bacterial Lipocalins: Origin, Structure, and Function. Austin (TX): Landes Bioscience. R. BishopC. CambillauG. PrivéD. HsiD. Tillo2000Bacterial Lipocalins: Origin, Structure, and Function.Austin (TX)Landes Bioscience
  22. 22. Skerra A (2000) Lipocalins as a scaffold. Biochim Biophys Acta 1482: 337–350.A. Skerra2000Lipocalins as a scaffold.Biochim Biophys Acta1482337350
  23. 23. Schiefner A, Chatwell L, Breustedt DA, Skerra A (2010) Structural and biochemical analyses reveal a monomeric state of the bacterial lipocalin Blc. Acta Crystallogr D Biol Crystallogr 66: 1308–1315.Schiefner A, Chatwell L, Breustedt DA, Skerra A2010Structural and biochemical analyses reveal a monomeric state of the bacterial lipocalin Blc.Acta Crystallogr D Biol Crystallogr6613081315
  24. 24. Bishop RE (2000) The bacterial lipocalins. Biochim Biophys Acta 1482: 73–83.RE Bishop2000The bacterial lipocalins.Biochim Biophys Acta14827383
  25. 25. UniProt Consortium (2010) The Universal Protein Resource (UniProt) in 2010. Nucleic Acids Res 38: D142–148.Consortium UniProt2010The Universal Protein Resource (UniProt) in 2010.Nucleic Acids Res38D142148
  26. 26. Campanacci V, Bishop RE, Blangy S, Tegoni M, Cambillau C (2006) The membrane bound bacterial lipocalin Blc is a functional dimer with binding preference for lysophospholipids. FEBS Lett 580: 4877–4883.V. CampanacciRE BishopS. BlangyM. TegoniC. Cambillau2006The membrane bound bacterial lipocalin Blc is a functional dimer with binding preference for lysophospholipids.FEBS Lett58048774883
  27. 27. Ganfornina MD, Gutierrez G, Bastiani M, Sanchez D (2000) A phylogenetic analysis of the lipocalin protein family. Mol Biol Evol 17: 114–126.MD GanforninaG. GutierrezM. BastianiD. Sanchez2000A phylogenetic analysis of the lipocalin protein family.Mol Biol Evol17114126
  28. 28. Woese CR (1987) Bacterial evolution. Microbiol Rev 51: 221–271.CR Woese1987Bacterial evolution.Microbiol Rev51221271
  29. 29. Vernikos GS, Parkhill J (2006) Interpolated variable order motifs for identification of horizontally acquired DNA: revisiting the Salmonella pathogenicity islands. Bioinformatics 22: 2196–2203.GS VernikosJ. Parkhill2006Interpolated variable order motifs for identification of horizontally acquired DNA: revisiting the Salmonella pathogenicity islands.Bioinformatics2221962203
  30. 30. Schlehuber S, Skerra A (2002) Tuning ligand affinity, specificity, and folding stability of an engineered lipocalin variant – a so-called ‘anticalin’ – using a molecular random approach. Biophys Chem 96: 213–228.S. SchlehuberA. Skerra2002Tuning ligand affinity, specificity, and folding stability of an engineered lipocalin variant – a so-called ‘anticalin’ – using a molecular random approach.Biophys Chem96213228
  31. 31. Schonfeld D, Matschiner G, Chatwell L, Trentmann S, Gille H, et al. (2009) An engineered lipocalin specific for CTLA-4 reveals a combining site with structural and conformational features similar to antibodies. Proc Natl Acad Sci U S A 106: 8198–8203.D. SchonfeldG. MatschinerL. ChatwellS. TrentmannH. Gille2009An engineered lipocalin specific for CTLA-4 reveals a combining site with structural and conformational features similar to antibodies.Proc Natl Acad Sci U S A10681988203
  32. 32. Kim HJ, Eichinger A, Skerra A (2009) High-affinity recognition of lanthanide(III) chelate complexes by a reprogrammed human lipocalin 2. J Am Chem Soc 131: 3565–3576.HJ KimA. EichingerA. Skerra2009High-affinity recognition of lanthanide(III) chelate complexes by a reprogrammed human lipocalin 2.J Am Chem Soc13135653576
  33. 33. Mills JL, Liu G, Skerra A, Szyperski T (2009) NMR structure and dynamics of the engineered fluorescein-binding lipocalin FluA reveal rigidification of beta-barrel and variable loops upon enthalpy-driven ligand binding. Biochemistry 48: 7411–7419.JL MillsG. LiuA. SkerraT. Szyperski2009NMR structure and dynamics of the engineered fluorescein-binding lipocalin FluA reveal rigidification of beta-barrel and variable loops upon enthalpy-driven ligand binding.Biochemistry4874117419
  34. 34. Acton TB, Gunsalus KC, Xiao R, Ma LC, Aramini J, et al. (2005) Robotic cloning and Protein Production Platform of the Northeast Structural Genomics Consortium. Methods Enzymol 394: 210–243.TB ActonKC GunsalusR. XiaoLC MaJ. Aramini2005Robotic cloning and Protein Production Platform of the Northeast Structural Genomics Consortium.Methods Enzymol394210243
  35. 35. Xiao R, Anderson S, Aramini J, Belote R, Buchwald WA, et al. (2010) The high-throughput protein sample production platform of the Northeast Structural Genomics Consortium. J Struct Biol 172: 21–33.R. XiaoS. AndersonJ. AraminiR. BeloteWA Buchwald2010The high-throughput protein sample production platform of the Northeast Structural Genomics Consortium.J Struct Biol1722133
  36. 36. Jansson M, Li YC, Jendeberg L, Anderson S, Montelione BT, et al. (1996) High-level production of uniformly 15N- and 13C-enriched fusion proteins in Escherichia coli. J Biomol NMR 7: 131–141.M. JanssonYC LiL. JendebergS. AndersonBT Montelione1996High-level production of uniformly 15N- and 13C-enriched fusion proteins in Escherichia coli.J Biomol NMR7131141
  37. 37. Neri D, Szyperski T, Otting G, Senn H, Wüthrich K (1989) Stereospecific nuclear magnetic resonance assignments of the methyl groups of valine and leucine in the DNA-binding domain of the 434 repressor by biosynthetically directed fractional 13C labeling. Biochemistry 28: 7510–7516.D. NeriT. SzyperskiG. OttingH. SennK. Wüthrich1989Stereospecific nuclear magnetic resonance assignments of the methyl groups of valine and leucine in the DNA-binding domain of the 434 repressor by biosynthetically directed fractional 13C labeling.Biochemistry2875107516
  38. 38. Kim S, Szyperski T (2003) GFT NMR, a new approach to rapidly obtain precise high-dimensional NMR spectral information. J Am Chem Soc 125: 1385–1393.S. KimT. Szyperski2003GFT NMR, a new approach to rapidly obtain precise high-dimensional NMR spectral information.J Am Chem Soc12513851393
  39. 39. Atreya HS, Szyperski T (2004) G-matrix Fourier transform NMR spectroscopy for complete protein resonance assignment. Proc Natl Acad Sci U S A 101: 9642–9647.HS AtreyaT. Szyperski2004G-matrix Fourier transform NMR spectroscopy for complete protein resonance assignment.Proc Natl Acad Sci U S A10196429647
  40. 40. Shen Y, Atreya HS, Liu G, Szyperski T (2005) G-matrix Fourier transform NOESY-based protocol for high-quality protein structure determination. J Am Chem Soc 127: 9085–9099.Y. ShenHS AtreyaG. LiuT. Szyperski2005G-matrix Fourier transform NOESY-based protocol for high-quality protein structure determination.J Am Chem Soc12790859099
  41. 41. Penhoat CH, Li Z, Atreya HS, Kim S, Yee A, et al. (2005) NMR solution structure of Thermotoga maritima protein TM1509 reveals a Zn-metalloprotease-like tertiary structure. J Struct Funct Genomics 6: 51–62.CH PenhoatZ. LiHS AtreyaS. KimA. Yee2005NMR solution structure of Thermotoga maritima protein TM1509 reveals a Zn-metalloprotease-like tertiary structure.J Struct Funct Genomics65162
  42. 42. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, et al. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6: 277–293.F. DelaglioS. GrzesiekGW VuisterG. ZhuJ. Pfeifer1995NMRPipe: a multidimensional spectral processing system based on UNIX pipes.J Biomol NMR6277293
  43. 43. Bartels C, Xia T-h, Billeter M, Güntert P, Wüthrich K (1995) The program XEASY for computer-supported NMR spectral analysis of biological macromolecules. Journal of Biomolecular NMR 6: 1–10.C. BartelsT-h XiaM. BilleterP. GüntertK. Wüthrich1995The program XEASY for computer-supported NMR spectral analysis of biological macromolecules.Journal of Biomolecular NMR6110
  44. 44. Zimmerman DE, Kulikowski CA, Huang Y, Feng W, Tashiro M, et al. (1997) Automated analysis of protein NMR assignments using methods from artificial intelligence. J Mol Biol 269: 592–610.DE ZimmermanCA KulikowskiY. HuangW. FengM. Tashiro1997Automated analysis of protein NMR assignments using methods from artificial intelligence.J Mol Biol269592610
  45. 45. Ulrich EL, Akutsu H, Doreleijers JF, Harano Y, Ioannidis YE, et al. (2008) BioMagResBank. Nucleic Acids Res 36: D402–408.EL UlrichH. AkutsuJF DoreleijersY. HaranoYE Ioannidis2008BioMagResBank.Nucleic Acids Res36D402408
  46. 46. Cornilescu G, Delaglio F, Bax A (1999) Protein backbone angle restraints from searching a database for chemical shift and sequence homology. J Biomol NMR 13: 289–302.G. CornilescuF. DelaglioA. Bax1999Protein backbone angle restraints from searching a database for chemical shift and sequence homology.J Biomol NMR13289302
  47. 47. Güntert P, Mumenthaler C, Wüthrich K (1997) Torsion angle dynamics for NMR structure calculation with the new program DYANA. J Mol Biol 273: 283–298.P. GüntertC. MumenthalerK. Wüthrich1997Torsion angle dynamics for NMR structure calculation with the new program DYANA.J Mol Biol273283298
  48. 48. Herrmann T, Güntert P, Wüthrich K (2002) Protein NMR structure determination with automated NOE assignment using the new software CANDID and the torsion angle dynamics algorithm DYANA. J Mol Biol 319: 209–227.T. HerrmannP. GüntertK. Wüthrich2002Protein NMR structure determination with automated NOE assignment using the new software CANDID and the torsion angle dynamics algorithm DYANA.J Mol Biol319209227
  49. 49. Liu G, Shen Y, Atreya HS, Parish D, Shao Y, et al. (2005) NMR data collection and analysis protocol for high-throughput protein structure determination. Proc Natl Acad Sci U S A 102: 10487–10492.G. LiuY. ShenHS AtreyaD. ParishY. Shao2005NMR data collection and analysis protocol for high-throughput protein structure determination.Proc Natl Acad Sci U S A1021048710492
  50. 50. Huang YJ, Tejero R, Powers R, Montelione GT (2006) A topology-constrained distance network algorithm for protein structure determination from NOESY data. Proteins 62: 587–603.YJ HuangR. TejeroR. PowersGT Montelione2006A topology-constrained distance network algorithm for protein structure determination from NOESY data.Proteins62587603
  51. 51. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, et al. (1998) Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr 54: 905–921.AT BrungerPD AdamsGM CloreWL DeLanoP. Gros1998Crystallography & NMR system: A new software suite for macromolecular structure determination.Acta Crystallogr D Biol Crystallogr54905921
  52. 52. Bhattacharya A, Tejero R, Montelione GT (2007) Evaluating protein structures determined by structural genomics consortia. Proteins 66: 778–795.A. BhattacharyaR. TejeroGT Montelione2007Evaluating protein structures determined by structural genomics consortia.Proteins66778795
  53. 53. Huang YJ, Powers R, Montelione GT (2005) Protein NMR recall, precision, and F-measure scores (RPF scores): structure quality assessment measures based on information retrieval statistics. J Am Chem Soc 127: 1665–1674.YJ HuangR. PowersGT Montelione2005Protein NMR recall, precision, and F-measure scores (RPF scores): structure quality assessment measures based on information retrieval statistics.J Am Chem Soc12716651674
  54. 54. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, et al. (2000) The Protein Data Bank. Nucleic Acids Res 28: 235–242.HM BermanJ. WestbrookZ. FengG. GillilandTN Bhat2000The Protein Data Bank.Nucleic Acids Res28235242
  55. 55. Petrey D, Fischer M, Honig B (2009) Structural relationships among proteins with different global topologies and their implications for function annotation strategies. Proc Natl Acad Sci U S A 106: 17377–17382.D. PetreyM. FischerB. Honig2009Structural relationships among proteins with different global topologies and their implications for function annotation strategies.Proc Natl Acad Sci U S A1061737717382
  56. 56. Petrey D, Honig B (2003) GRASP2: visualization, surface properties, and electrostatics of macromolecular structures and sequences. Methods Enzymol 374: 492–509.D. PetreyB. Honig2003GRASP2: visualization, surface properties, and electrostatics of macromolecular structures and sequences.Methods Enzymol374492509
  57. 57. Yang AS, Honig B (2000) An integrated approach to the analysis and modeling of protein sequences and structures. I. Protein structural alignment and a quantitative measure for protein structural distance. J Mol Biol 301: 665–678.AS YangB. Honig2000An integrated approach to the analysis and modeling of protein sequences and structures. I. Protein structural alignment and a quantitative measure for protein structural distance.J Mol Biol301665678
  58. 58. Hasegawa H, Holm L (2009) Advances and pitfalls of protein structural alignment. Curr Opin Struct Biol 19: 341–348.H. HasegawaL. Holm2009Advances and pitfalls of protein structural alignment.Curr Opin Struct Biol19341348
  59. 59. Koradi R, Billeter M, Wüthrich K (1996) MOLMOL: a program for display and analysis of macromolecular structures. J Mol Graph 14: 51–55, 29–32: R. KoradiM. BilleterK. Wüthrich1996MOLMOL: a program for display and analysis of macromolecular structures.J Mol Graph 14: 51–55,29–32
  60. 60. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402.SF AltschulTL MaddenAA SchafferJ. ZhangZ. Zhang1997Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.Nucleic Acids Res2533893402
  61. 61. Arnold K, Bordoli L, Kopp J, Schwede T (2006) The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22: 195–201.K. ArnoldL. BordoliJ. KoppT. Schwede2006The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling.Bioinformatics22195201
  62. 62. Kersey PJ, Lawson D, Birney E, Derwent PS, Haimel M, et al. (2010) Ensembl Genomes: extending Ensembl across the taxonomic space. Nucleic Acids Res 38: D563–569.PJ KerseyD. LawsonE. BirneyPS DerwentM. Haimel2010Ensembl Genomes: extending Ensembl across the taxonomic space.Nucleic Acids Res38D563569
  63. 63. Luthy R, Bowie JU, Eisenberg D (1992) Assessment of protein models with three-dimensional profiles. Nature 356: 83–85.R. LuthyJU BowieD. Eisenberg1992Assessment of protein models with three-dimensional profiles.Nature3568385
  64. 64. Laskowski RA, MacArthur MW, Moss DS, Thornton JM (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Cryst 26: 283–291.RA LaskowskiMW MacArthurDS MossJM Thornton1993PROCHECK: a program to check the stereochemical quality of protein structures.J Appl Cryst26283291
  65. 65. Sippl MJ (1993) Recognition of errors in three-dimensional structures of proteins. Proteins 17: 355–362.MJ Sippl1993Recognition of errors in three-dimensional structures of proteins.Proteins17355362
  66. 66. Bhattacharya A, Tejero R, Montelione GT (2007) Evaluating protein structures determined by structural genomics consortia. Proteins 66: 778–795.A. BhattacharyaR. TejeroGT Montelione2007Evaluating protein structures determined by structural genomics consortia.Proteins66778795