Anomer-Specific Recognition and Dynamics in a ... - ACS Publications

Feb 4, 2016 - with both the α- and the β-anomer of L-fucose and the “fucose ... assignment of the split signals to the α- or β-anomer was confir...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OSNABRUECK

Article

Anomer-Specific Recognition and Dynamics in a Fucose-binding Lectin Pawel M. Antonik, Alexander N Volkov, Ursula N. Broder, Daniele Lo Re, Nico A.J. Van Nuland, and Peter B. Crowley Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.5b01212 • Publication Date (Web): 04 Feb 2016 Downloaded from http://pubs.acs.org on February 10, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Anomer-Specific Recognition and Dynamics in a Fucose-binding Lectin Paweł M. Antonik,a,b,# Alexander N. Volkov,c,d,# Ursula N. Broder,a Daniele Lo Re,a Nico A. J. van Nuland,c,d and Peter B. Crowleya,*

a

School of Chemistry, National University of Ireland Galway, University Road, Galway,

Ireland. b

Department of Food BioSciences, Teagasc Food Research Centre Ashtown, Dublin 15,

Ireland. c

Jean Jeener NMR Centre, Structural Biology Brussels, Vrije Universiteit Brussel, Pleinlaan 2,

1050 Brussels, Belgium. d

Structural Biology Research Centre, VIB, Pleinlaan 2, 1050 Brussels, Belgium.

#

These authors contributed equally

*Correspondence to: [email protected], +353 91 49 24 80

Keywords: Anomer, Carbohydrates, Lectin, NMR Spectroscopy, HSQC

Funding Sources This research was supported by Teagasc (Walsh fellowship to PA), FWO (postdoctoral fellowship to ANV), Marie Curie Action (grant PIEF-GA-2011-299042 to DLR) the Hercules Foundation (infrastructure grant to Jean Jeener NMR centre), NUI Galway (NMR infrastructure support and Millennium Fund to PBC) and Science Foundation Ireland (grant 10/RFP/BIC2807 to PBC).

ACS Paragon Plus Environment

1

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Abstract Sugar binding by a cell surface ~29 kDa lectin (RSL) from the bacterium Ralstonia solanacearum was characterized by NMR spectroscopy. The complexes formed with four monosaccharides and four fucosides were studied. Complete resonance assignments and backbone dynamics were determined for RSL in the sugar-free form and when bound to Lfucose or D-mannose. RSL was found to interact with both the α- and the β-anomer of Lfucose and the “fucose like” sugars, D-arabinose and L-galactose. Peak splitting was observed for some resonances of the binding site residues. The assignment of the split signals to the αor β-anomer was confirmed by comparison with the spectra of RSL bound to methyl-α-Lfucoside or methyl-β-L-fucoside. The backbone dynamics of RSL were sensitive to the presence of ligand, with the protein adopting a more compact structure upon binding to Lfucose. Taking advantage of tryptophan residues in the binding sites, we show that the indole resonance is an excellent reporter on ligand binding. Each sugar resulted in a distinct signature of chemical shift perturbations – suggesting that tryptophan signals are a sufficient probe of sugar binding.

ACS Paragon Plus Environment

2

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Introduction Carbohydrate recognition is a topic of far-reaching importance1 that continues to challenge the protein scientist2–6 and the synthetic chemist7–10 alike. The principal challenge lies in understanding the balance of noncovalent interactions that favour receptor binding over solvation in water. Complex formation in the case of monosaccharides has the added feature of selection between the α and the β anomeric forms. NMR spectroscopy offers the powerful tools of chemical shift perturbation (∆δ) analysis to characterize protein-ligand binding. Here, we have applied this technique to study the fucose-binding lectin11 (RSL) from the bacterium Ralstonia solanacearum. Knowledge of fucose recognition by lectins is expected to aid the development of new strategies to combat bacterial disease. For example, host-pathogen interactions may be mediated by lectins that bind to fucosylated oligosaccharides present in plant cell walls or mammalian epithelial tissue.12 In the case of R. solanacearum, extracellular fucose-binding lectins likely contribute to infection and progression of wilt disease in potato and tomato.11 RSL is a ~29 kDa trimer with a 6-bladed β-propeller fold.11 Each monomer comprises two blades, corresponding to the N- and C-terminal halves of the protein, which share ~40 % sequence identity (Figure 1). There are two sugar-binding sites per monomer. Site 1 is intramonomeric and is nestled between the N- and C-terminal blades of the monomer. Site 2 is inter-monomeric and occurs between the blades of adjacent monomers (Figure 1). Each site contains a WiXGXGWi+5 motif. The indole ring of Wi forms CH-π bonds13,14 with the monosaccharide while the indole of Wi+5 donates a hydrogen bond to hydroxyl O3 of fucose.11 The occurrence of two almost identical binding sites makes RSL an interesting model protein for ∆δ analysis. The protein is exceptionally stable (Tm = 86 °C in the sugarfree form),15 which favours NMR studies. There are numerous crystal structures (including PDB 2BT9 at 0.94 Å resolution) of RSL11 and related lectins15–18 in sugar-free and -bound forms. Furthermore, ligand binding to RSL has been characterized by isothermal titration calorimetry11,14 and computational methods.14,19,20 The interactions of RSL with four monosaccharides and four fucosides were investigated by chemical shift perturbation studies on

15

N-labelled RSL. Remarkably, we

observed anomer-specific effects with resonance splitting in the presence of L-fucose or related monosaccharides. While many lectin-carbohydrate complexes have been characterized by HSQC spectroscopy4–6,21–30 we have not found evidence of such anomer-specific effects in the literature. NMR binding data for α- and β-methyl- or -nitrophenyl-L-fucosides together

ACS Paragon Plus Environment

3

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

with analysis of the RSL crystal structure suggest that a Tyr side chain that flanks the intramonomeric site may act as a steric impediment to the binding of β-L-fucosides. The backbone dynamics of the sugar-free and sugar-bound forms of RSL were characterized in the ns to s timescales. Our data indicate an increased rigidity of the RSL structure upon sugar binding, consistent with a ligand-induced fit. We show also that the tryptophan indole –NH resonance is an excellent probe of ligand binding.31,32 In the case of RSL with three Trp side chains per binding site, sugar-specific signatures were obtained.

ACS Paragon Plus Environment

4

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Experimental Section Materials. Carbohydrates were purchased from SIGMA (D-arabinose, L-fucose, L-galactose, D-mannose,) or Carbosynth (methyl-α-L-fucoside, methyl-β-L-fucoside, 4-nitrophenyl-α-Lfucoside and 4-nitrophenyl-β-L-fucoside).

13

C-labelled D-glucose and

15

(NH4)2SO4 were

purchased from Cortecnet.

Protein Production.

15

N- and

13

C/15N-RSL samples were produced in E. coli BL21

transformed with the plasmid pET25rsl.11 Cultures were grown to mid-log phase on LB and then transferred to minimal medium (supplemented with 75 mg/ml carbenicillin). For labeling the medium contained 1 g/L

15

(NH4)2SO4 as the sole nitrogen source. For

13

15

N-

C/15N-

labeling 2 g/l 13C-labeled glucose was added as the sole carbon source. RSL was purified by affinity chromatography (D-mannose-agarose resin, SIGMA) on an ÄKTA FPLC, as described previously. Extensive dialysis was required to remove D-mannose.11 The protein concentration was determined by UV spectroscopy using an extinction coefficient ε280 = 44.6 mM-1cm-1.

NMR Assignments. The samples contained 2 mM

13

C/15N-RSL in 20 mM potassium

phosphate, 50 mM NaCl pH 6.1 and 6 % D2O for the lock. The sugar-bound samples were prepared by the addition of D-mannose or L-fucose to a final concentration of 24 mM (i.e. two equivalents). NMR experiments were performed at 303 K on Varian NMR Direct-Drive System 600 and 800 MHz spectrometers, the latter equipped with a salt-tolerant 13C-enhanced PFG-Z cold probe. All NMR data were processed in NMRPipe33 and analysed in CCPN.34 The assignment of the backbone resonances were determined from a set of 2D 1H-15N HSQC and 3D HNCACB, CBCA(CO)NH, and HNCO experiments.35 The tryptophan indoles were assigned from a combination of 2D (HB)CB(CGCD)HD36 and 3D

15

N- and

13

C-NOESY-

HSQC spectra, that provide Cβ-Hδ correlations and strong intra-residue Hδ-Hε NOEs, respectively. Finally, the side chain NH2 resonances of asparagine and glutamine residues were assigned from the CBCA(CO)NH spectrum by correlating with the corresponding Cα/Cβ (Asn) or Cβ (Gln) chemical shifts. The assignments for RSL have been deposited in the BMRB with accession numbers 25950 (bound to D-mannose), 25951 (bound to L-fucose) and 25952 (sugar-free). Sugar Binding and Chemical Shift Perturbations. Samples contained 0.25 mM 15N-RSL in 20 mM potassium phosphate, 50 mM NaCl pH 6.0 and 10 % D2O. Monosaccharides or

ACS Paragon Plus Environment

5

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

fucosides were added to a final concentration of 3 mM. Spectra were acquired on sugar-free RSL and sugar-bound RSL samples at 303 K on an Agilent 600 MHz NMR spectrometer equipped with a HCN cold probe. The differences in chemical shifts (∆δ) between the sugarfree and the sugar-bound RSL were measured in CCPN34 and the average perturbations were calculated as ∆δavg = {[∆δH2 + (0.2 × ∆δN2)]/2}½.37 NMR Dynamics. Data were acquired on samples of sugar-free RSL and RSL bound to Lfucose or D-mannose. The backbone

15

N R1, R2, CPMG R2ex relaxation dispersion, and 15N

ZZ-exchange experiments were recorded on a 600 MHz spectrometer at 303 K. The

15

N R1

and R2 relaxation rates were obtained from a series of 2D experiments with coherence selection achieved by pulse field gradients.38 The relaxation parameters and corresponding errors were extracted with CCPN,34 and the transverse relaxation rates were corrected for the R2ex contribution (see below). The diffusion tensors of the free and bound RSL were obtained from the experimental R2/R1 values using the programme r2r1_diffusion (A. Palmer). Residues with a

15

N R2 higher than one standard deviation from the average (obtained

for the entire set of the backbone amides) likely exhibit significant internal motions39 and were excluded from the analysis. The model selection was performed using F statistics as described elsewhere.40 The CPMG relaxation dispersion experiments41 were recorded with 0, 25, 50 (in duplicate), 75, 125, 175, 225, 350, 550, 750 (in duplicate), and 1000 Hz pulse repetition rates. The peak intensities at each repetition rate were determined in CCPN, and the subsequent analysis was performed in NESSY,42 providing the R2ex values and their errors. The

15

N ZZ-exchange experiments43 were acquired as a series of interleaved 2D

spectra with mixing times of 0 (in duplicate), 20, 50 (in duplicate), 90, 150, 230, 340, 500, 720, 1030, and 1500 ms. The ZZ exchange intensity build-up curves were analyzed using a composite ratio of four peak intensities (two auto-peaks, AA and BB, and two cross-peaks, AB and BA) according to Eq. 1:44

Ξ =

   =  ≅    1    −   

where I(t) is the corresponding peak intensity at the mixing time t and kon and koff are, respectively, the association and dissociation rate constants for the exchange process. A good fit (χ2red = 0.00069) was obtained for the A85 data, with ζ = 1.235 ± 0.022 s-2. For a system in

ACS Paragon Plus Environment

6

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

a two-state A ↔ B equilibrium, the equilibrium constant is given by K = [A]/[B] = k1/k-1, where k1 and k-1 are first-order association and dissociation rate constants, respectively, and the relative populations of the two species, [A]/[B], can be estimated from the ratio of the corresponding peak intensities in the fully-relaxed HSQC spectrum.44 With K = [A]/[B] = k1/k-1 = 0.90 ± 0.12 for the A85 backbone amide resonances of RSL bound to the α- or βanomers of L-fucose and ζ = k1k-1 = 1.235 ± 0.022 s-2, we obtain k1 = 1.05 ± 0.16 s-1 and k-1 = 1.17 ± 0.20 s-1, yielding the final value of the exchange rate constant kex = k1 + k-1 = 2.22 ± 0.25 s-1.

ACS Paragon Plus Environment

7

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Results and Discussion Backbone Resonance Assignments. The 1H-15N HSQC spectrum, with a total of 105 wellresolved cross-peaks, confirmed that RSL is a symmetric trimer in solution. Despite the relatively high molecular weight (29.2 kDa) the protein yielded an excellent HSQC comparable to that of a 10 kDa protein. Almost complete 1HN, 15N, C′, Cα and Cβ assignments were determined for sugar-free RSL. Figure S1A shows the assigned 1H-15N HSQC spectrum. Pairs of homologous residues, due to the sequence and structural similarity of the N- and Cterminal sub-domains, had resonances with closely similar chemical shifts and line widths (Figure S1A; S9/S52, I16/I61, R17/R62, W31/W76, D32/D77, W36/W81, and A40/A85, F41/Y86, see also Figure 2). Interestingly, no peaks were observed for the glycines that occur in the WiXGXGWi+5 motifs (residues 31-36 and 76-81, Figure 1), suggestive of exchange broadening for these turn residues.

Chemical Shift Perturbation due to Sugar Binding. Monosaccharide/fucoside binding to RSL was in slow-exchange on the NMR time scale. The presence of D-arabinose, L-fucose, methyl-α-L-fucoside,

methyl-β-L-fucoside,

nitrophenyl-α-L-fucoside,

nitrophenyl-β-L-

fucoside, L-galactose or D-mannose (with Kd values ranging from < 1 - 100 µM11,19), resulted in distinct perturbations of the backbone 1HN, 15N resonances (Figures 3). Complete backbone assignments were determined for RSL bound to L-fucose or D-mannose (see Figures S1B and S1C, respectively for the assigned HSQCs). In the case of the other sugars, only the 1H-15N HSQC spectrum was assigned. The sugars D-arabinose, L-fucose, and L-galactose have identical stereochemistry and differ only in the nature of the C5 substituent; –H, –CH3 and –CH2OH, respectively (Figure 3). In the crystal structure of RSL bound to methyl-α-L-fucoside11 the C6 methyl is inserted into a shallow hydrophobic pocket formed by the side chains of Trp10, Ile59, Ile61, and Trp76 in the intra-monomeric site or Trp53, Pro14, Ile16, and Trp31 in the inter-monomeric site (Figure 1). In addition to interactions with these side chains the methyl substituent is also in van der Waals contact with two water molecules. The nature of this portion of the binding site, in particular, the water accessibility suggests a molecular basis for the broad specificity of RSL. In the case of D-arabinose the absence of the C6 methyl is likely compensated by the insertion of a water molecule (Figure 4B). When L-galactose binds, its hydroxyl substituent can orient towards the solvent and there are no unfavourable clashes. The tolerance of this

ACS Paragon Plus Environment

8

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

hydrophobic pocket to polar substituents is evidenced by the binding of D-mannose. By virtue of their similar stereochemistry, L-fucose and D-mannose can bind to the same lectins.11,16,45 Specifically, the protein-sugar contacts mediated by hydroxyls O2, O3 and O4 of L-fucose can be established likewise by the O4, O3 and O2 groups of D-mannose (Figure 3). Consequently, when D-mannose binds to RSL its anomeric hydroxyl is inserted into the hydrophobic pocket. And considering the stereochemistry of C5 in L-fucose it is apparent that RSL must bind β-D-mannose. One might also speculate that the ~50 fold lower affinity for Dmannose (compared to L-fucose) is the result of a weakened hydrophobic effect at this site. In the presence of D-arabinose, L-fucose or L-galactose, the number of resonances in 1

15

the H- N HSQC spectrum of RSL increased by ~10 %. The assignment of RSL bound to Lfucose revealed that the backbone amide resonances of six residues (Y37, F41, K83, G84, A85 and Y86) were split into two clearly resolved peaks. These pairs of resonances were attributed to the binding of the α and β forms of L-fucose, which is present in solution in a ratio of 0.3 α-L-fucose to 0.7 β-L-fucose.46 When bound to RSL, L-fucose is positioned with the anomeric hydroxyl exposed to the solvent.11,15 Thus, the α and β forms can bind with minimal steric clashes between the protein and the anomeric hydroxyl. Crystallographic evidence for this effect was reported in a RSL crystal structure (PDB 4I6S), where both anomers of L-fucose were bound.15 As such RSL is suited to binding fucose-terminated oligosaccharides. While the lack of α/β anomer discrimination or selection is known for lectins,16,17,47 we have not found NMR evidence for this effect in the literature. The assignment of the split peaks to the binding of the α and β forms was confirmed by comparison with the spectra of RSL in complex with methyl-α-L-fucoside or methyl-β-Lfucoside (Figure 2). There was good overlap between the fucose-split resonances and the single resonance in the fucoside complex. Note that the complex of RSL with D-mannose did not result in peak splitting in the HSQC. Only the β-D-mannose can bind for reasons already discussed. Pronounced chemical shift changes (∆δavg = 0.4-0.6 ppm) were measured for the resonances of equivalent residues Ala40 and Ala85 (Figure 2). These large effects can be rationalized by comparison with the X-ray structure.11 The amide NH of Ala40 and Ala85 donate a hydrogen bond to hydroxyl O2 of methyl-α-L-fucoside (N---O2, 2.9 Å) in the intraand inter-monomeric binding sites, respectively.

ACS Paragon Plus Environment

9

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

Tryptophan Indole Probes of Sugar Binding. The pattern of backbone chemical shift perturbations was similar for D-arabinose, L-fucose, L-galactose and D-mannose (Figure 3). Despite the differences in chemical structure, the main interactions (e.g. hydrogen bonds and CH-π bonds) were maintained by each sugar and the backbone resonances of RSL sensed similar changes to the chemical environment. In contrast, the –NH resonance of the tryptophan indole was a more sensitive reporter,31,32 and a unique pattern of shifts was observed for each monosaccharide. There are seven tryptophans in RSL, three in the intramonomeric site (Trp10, Trp76 and Trp81), and three in the inter-monomeric (Trp53, Trp31 and Trp36). The seventh tryptophan (Trp74) does not contribute directly to either site. Figure 4A shows the indole –NH perturbations in the presence of the four sugars. As discussed, the differences in binding for this set of sugars were focused on the C5 substituent (or C1 in the case of D-mannose, Figure 3). In the crystal structure, the methyl C6 of methylα-L-fucoside makes van der Waals contacts with the indoles of Trp 31/76 and Trp53/10. The shortest point of contact with C6 is to the indole –NH in the Trp31/76 pair (Figure 4B). Therefore, it could be expected that these indole resonances will have the largest differences in perturbation when the various sugars bind. This was clearly the case (Figure 4A, yellow bars). The Trp36/81 pair is furthest from C5, and the indole –NH of these side chains makes a hydrogen bond to hydroxyl O3, which is likely to dominate the chemical shift. Consistent with this structural interpretation, the Trp36/81 resonances had similar, large perturbations irrespective of which sugar was bound (Figure 4A, red bars). The Trp53/10 pair had similar perturbations for D-arabinose, L-fucose and L-galactose, while shifts of similar size but opposite sign were observed for D-mannose (Figure 4A, blue bars). Finally, the indole resonance of Trp74 had negligible chemical shift perturbations as this group was not involved in sugar binding. As with the backbone amide resonances, some of the indole resonances were also split due to the binding of α- or β-L-fucose (Figure 2C and 4A). Again the assignment of the split signals was confirmed by comparison with the data for methyl-α- or methyl-β-Lfucoside. Considering the sugar-specific signatures observed for the indole resonances (versus the similar pattern of backbone amide shifts), the use of

15

N-tryptophan labelled31 samples

(rather than uniform labelling) may provide a simpler probe for the characterization of sugar binding. An effective application would require two or more binding site tryptophans with structurally distinct ligand interactions.

ACS Paragon Plus Environment

10

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Steric Effects in the Binding Site. The observation of distinct chemical shift perturbations for the α- and β-anomers of L-fucose and methyl-L-fucoside prompted further analysis of this effect. In the atomic resolution crystal structure of RSL11 (PDB 2BT9) the methyl substituent of methyl-α-L-fucoside is solvent exposed (Figure 1B, right panel). The solvent exposure of the anomeric substituent implies that β-L-fucosides can be accommodated. However, the side chain of Tyr37 adjacent to the intra-monomeric binding site may impede the binding of bulky β-L-fucosides. The corresponding residue in the inter-monomeric site is Thr82 (Figure 1B). Figures 5 and S2 show overlaid spectral regions of RSL bound to nitrophenyl-α-Lfucoside, nitrophenyl-β-L-fucoside or the corresponding methyl-L-fucoside. In some cases the same chemical shift perturbation (with respect to sugar-free RSL) was observed for the α- and β-, methyl- or nitrophenyl-L-fucoside. For other resonances, the β-L-fucosides gave rise to perturbations that differed significantly from the α-L-fucoside shifts. For example, the R62 resonance was similarly altered in the presence of all four fucosides while the equivalent R17 in the intra-monomeric site was not shifted by the β-anomers. While it is difficult to pinpoint chemical shift changes to specific structural effects (due to the proximity of the intra- and inter-monomeric sites) it seems reasonable to assume that steric interactions between the βsubstituent and the Tyr37 side chain (and the lack of such an interaction at Thr82) results in a different contribution to the chemical shift perturbations of binding site resonances. A role for tyrosine side chains in lectin selectivity has been proposed previously, for example, the tyrosine gate in fimH.5,6 Perhaps, the single Tyr37 serves to modulate the binding affinity of the three intra-monomeric sites in RSL.

Dynamics of Sugar-free and Sugar-bound RSL. The dynamics of the RSL backbone on the ps-ns timescale were assessed by using the residue-based squared-order parameter (S2).48 With an average S2 = 0.82 ± 0.08, the sugar-free RSL backbone was rigid, except for the homologous loop regions comprising residues 32-35 and 77-80 (Figure 6, upper panels and compare Figure 1). These considerably flexible (S2 < 0.75) loops contain two glycines (no amide resonances detected) and connect two of the sugar-binding tryptophans in the intra- and inter-monomeric binding sites. The flexibility of these loops was not affected by binding to either L-fucose or D-mannose. Residues 40-44 and 64-72 exhibited minor changes in the order parameters (|∆S2| < 0.05), while S2 for the rest of the backbone remained constant. These data suggest that sugar binding did not affect the RSL backbone dynamics on the ps-ns timescale.

ACS Paragon Plus Environment

11

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

To complement the S2 analysis,

15

Page 12 of 29

N R1 and R2 relaxation parameters were also

measured. Governed by the rotational diffusion of a molecule, the R1 and R2 relaxation rates can provide information on the hydrodynamic properties of the protein. A small, uniform decrease in the R2/R1 ratios was observed for the sugar-bound forms of RSL compared to sugar-free RSL (Figure 6, middle panels). Quantitative analysis of the R2/R1 profiles yielded isotropic diffusion tensors with rotational correlation time τc = 12.42 or 11.75 ns for the sugar-free and -bound RSL, respectively.49 These data are consistent with an RSL trimer in solution and suggest that the protein adopts a more compact structure upon sugar binding. RSL dynamics were probed further by Carr-Purcell-Meiboom-Gill (CPMG) relaxation dispersion spectroscopy.50,51 In this experiment, protein groups that undergo conformational exchange on the µs-ms timescale give rise to relaxation dispersion curves, which provide an exchange contribution (R2ex) to the transverse relaxation rate R2. In sugar-free RSL, ~40 % of residues experienced R2ex > 0 (Figure 6, lower panels), suggesting that the protein was in conformational exchange between two or more lowly-populated species. The clustering of these exchange effects suggests the possibility of concerted motions. Mannose binding moderately altered the distribution of the R2ex terms, with a reduction in the number of residues with R2ex contributions. In contrast, fucose binding abolished the R2ex terms for most of the protein. Such a drastic decrease in R2ex values suggests that the conformational dynamics observed in sugar-free RSL were quenched upon fucose binding. These data, together with the τc results, indicate an increased rigidity of the sugar-bound protein.1,21,27 Finally, by using

15

N ZZ-exchange spectroscopy,51 we studied the slow dynamics of

RSL bound to the α- and β-anomer of L-fucose. In this experiment, magnetization arising from species in slow exchange (1 s-1 ≤ kex ≤ 103 s-1) gives rise to symmetry-related crosspeaks. The build-up of the cross-peak intensities, followed in a series of spectra recorded with varying mixing time, can be used to extract the kinetic parameters. At long mixing times (0.5 - 1 s), the resonances of RSL bound to α- or β-L-fucose featured distinct cross-peaks (Figure 7), indicating that the two bound forms were in exchange. Only the A85 resonances were resolved sufficiently for quantitative analysis (Figure 7C) and an exchange constant kex = 2.22 ± 0.25 s-1 was obtained (see Experimental Section). Consistent with the µM binding affinity11 the NMR data suggest that the α- and β-L-fucose-bound forms of RSL undergo very slow exchange. The dynamics of sugar-binding proteins have been assessed in numerous studies, some of which have pointed to a role for conformational entropy.5,6,24,25 In some examples, it was found that the sugar-bound form was more dynamic than the sugar-free form, with the

ACS Paragon Plus Environment

12

Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

entropy gain being thermodynamically favorable for ligand binding.25 In the case of RSL, the protein dynamics were reduced in the sugar-bound form. Upon binding to L-fucose the protein became more compact (decreased τc) and motions on the µs-ms timescale were frozen out. Interestingly, the fast dynamics analysis revealed flexible loops (S2 < 0.75) adjacent to both the intra- and inter-monomeric binding sites. Here, the flexibility was not affected by binding to monosaccharides. These loops extend away from the binding site and may play a structural role in RSL interactions with cell surface oligosaccharides.

ACS Paragon Plus Environment

13

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

Conclusions Using NMR spectroscopy, we have shown that the bacterial lectin RSL binds both the α- and the β-anomer of L-fucose and “fucose like” sugars. Anomer-specific chemical shift perturbations were observed for methyl- and nitrophenyl-substituted L-fucosides. A 15N ZZexchange experiment confirmed that RSL binds to α- or β-L-fucose, which exchange on the second timescale. This appears to be the first example of such anomer-specific interactions measured by NMR. Analysis of the atomic resolution crystal structure suggests further that the accessibility of the intra- and inter-monomeric binding sites is different. A Tyr side chain (Tyr37) adjacent to the intra-monomeric may act as a steric impediment to the binding of β-Lfucosides with bulky substituents. The relaxation data reveal that RSL adopts a more compact structure when bound to L-fucose, suggesting a ligand-induced fit. The protein compaction due to sugar-binding was accompanied by decreased backbone motions on the µs-ms timescale. Finally, in addition to the conventional analysis of backbone amide chemical shifts we show that the tryptophan indole resonance is an excellent reporter of ligand binding, with characteristic ∆δ signatures observed for each sugar.

ACS Paragon Plus Environment

14

Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Acknowledgements We thank M. Wimmerová (Masaryk University) for providing the pET25rsl plasmid and E. Tishchenko (Agilent Technologies) for setting up the ZZ exchange and relaxation dispersion experiments.

ACS Paragon Plus Environment

15

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

References

(1) Gabius, H.-J., Andre, S., Jimenez-Barbero, J., Romero, A., and Solis, D. (2011) From lectin structure to functional glycomics: principles of the sugar code. Trends Biochem. Sci. 36, 298–313. (2) Fiege, B., Rademacher, C., Cartmell, J., Kitov, P. I., Parra, F., and Peters, T. (2012) Molecular Details of the Recognition of Blood Group Antigens by a Human Norovirus as Determined by STD NMR Spectroscopy. Angew. Chemie Int. Ed. 51, 928–932. (3) de Bentzmann, S., Varrot, A., and Imberty, A. (2014) Monitoring Lectin Interactions with Carbohydrates. Methods Mol. Biol. 1149, 403–414. (4) Pederson, K., Mitchell, D. A., and Prestegard, J. H. (2014) Structural Characterization of the DC-SIGN–LewisX Complex. Biochemistry 53, 5700–5709. (5) Vanwetswinkel, S., Volkov, A. N., Sterckx, Y. G. J., Garcia-Pino, A., Buts, L., Vranken, W. F., Bouckaert, J., Roy, R., Wyns, L., and van Nuland, N. A. J. (2014) Study of the Structural and Dynamic Effects in the FimH Adhesin upon α-d-Heptyl Mannose Binding. J. Med. Chem. 57, 1416–1427. (6) Fiege, B., Rabbani, S., Preston, R. C., Jakob, R. P., Zihlmann, P., Schwardt, O., Jiang, X., Maier, T., and Ernst, B. (2015) The Tyrosine Gate of the Bacterial Lectin FimH: A Conformational Analysis by NMR Spectroscopy and X-ray Crystallography. ChemBioChem 16, 1235–1246. (7) Segura, M., Bricoli, B., Casnati, A., Muñoz, E. M., Sansone, F., Ungaro, R., and Vicent, C. (2003) A Prototype Calix[4]arene-Based Receptor for Carbohydrate Recognition Containing Peptide and Phosphate Binding Groups. J. Org. Chem. 68, 6296–6303. (8) Ferrand, Y., Crump, M. P., and Davis, A. P. (2007) A synthetic lectin analog for biomimetic disaccharide recognition. Science (80). 318, 619–622. (9) Jang, Y., Natarajan, R., Ko, Y. H., and Kim, K. (2014) Cucurbit[7]uril: A High-Affinity Host for Encapsulation of Amino Saccharides and Supramolecular Stabilization of Their αAnomers in Water. Angew. Chemie Int. Ed. 53, 1003–1007. (10) Chandramouli, N., Ferrand, Y., Lautrette, G., Kauffmann, B., Mackereth, C. D., Laguerre, M., Dubreuil, D., and Huc, I. (2015) Iterative design of a helically folded aromatic oligoamide sequence for the selective encapsulation of fructose. Nat. Chem. 7, 334–341. (11) Kostlanova, N., Mitchell, E. P., Lortat-Jacob, H., Oscarson, S., Lahmann, M., GilboaGarber, N., Chambat, G., Wimmerova, M., and Imberty, A. (2005) The fucose-binding lectin from Ralstonia solanacearum - A new type of beta-propeller architecture formed by

ACS Paragon Plus Environment

16

Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

oligomerization and interacting with fucoside, fucosyllactose, and plant xyloglucan. J. Biol. Chem. 280, 27839–27849. (12) Esko, J. D., and Sharon, N. (2009) Microbial Lectins: Hemagglutinins, Adhesins, and Toxins, in Essentials of Glycobiology. (Varki, A., Cummings, R. D., Esko, J. D., Freeze, H. H., Stanley, P., Bertozzi, C. R., Hart, G. W., and Etzler, M. E., Eds.) 2nd ed., pp 489–500. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press. (13) Asensio, J. L., Arda, A., Canada, F. J., and Jimenez-Barbero, J. (2013) CarbohydrateAromatic Interactions. Acc. Chem. Res. 46, 946–954. (14) Wimmerova, M., Kozmon, S., Necasova, I., Mishra, S. K., Komarek, J., and Koca, J. (2012) Stacking Interactions between Carbohydrate and Protein Quantified by Combination of Theoretical and Experimental Methods. PLoS One 7. (15) Arnaud, J., Claudinon, J., Tröndle, K., Trovaslet, M., Larson, G., Thomas, A., Varrot, A., Römer, W., Imberty, A., and Audfray, A. (2013) Reduction of Lectin Valency Drastically Changes Glycolipid Dynamics in Membranes but Not Surface Avidity. ACS Chem. Biol. 8, 1918–1924. (16) Wimmerova, M., Mitchell, E., Sanchez, J. F., Gautier, C., and Imberty, A. (2003) Crystal structure of fungal lectin - Six-bladed beta-propeller fold and novel fucose recognition mode for Aleuria aurantia lectin. J. Biol. Chem. 278, 27059–27067. (17) Fujihashi, M., Peapus, D. H., Kamiya, N., Nagata, Y., and Miki, K. (2003) Crystal Structure of Fucose-Specific Lectin from Aleuria aurantia Binding Ligands at Three of Its Five Sugar Recognition Sites. Biochemistry 42, 11093–11099. (18) Audfray, A., Claudinon, J., Abounit, S., Ruvoen-Clouet, N., Larson, G., Smith, D. F., Wimmerova, M., Le Pendu, J., Romer, W., Varrot, A., and Imberty, A. (2012) Fucosebinding Lectin from Opportunistic Pathogen Burkholderia ambifaria Binds to Both Plant and Human Oligosaccharidic Epitopes. J. Biol. Chem. 287, 4335–4347. (19) Mishra, S. K., Adam, J., Wimmerova, M., and Koca, J. (2012) In Silico Mutagenesis and Docking Study of Ralstonia solanacearum RSL Lectin: Performance of Docking Software To Predict Saccharide Binding. J. Chem. Inf. Model. 52, 1250–1261. (20) Mishra, S. K., Calabró, G., Loeffler, H. H., Michel, J., and Koča, J. (2015) Evaluation of Selected Classical Force Fields for Alchemical Binding Free Energy Calculations of ProteinCarbohydrate Complexes. J. Chem. Theory Comput. 11, 3333–3345. (21) Canales-Mayordomo, A., Fayos, R., Angulo, J., Ojeda, R., Martin-Pastor, M., Nieto, P. M., Martin-Lomas, M., Lozano, R., Gimenez-Gallego, G., and Jimenez-Barbero, J. (2006) Backbone dynamics of a biologically active human FGF-1 monomer, complexed to a

ACS Paragon Plus Environment

17

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

hexasaccharide heparin-analogue, by N-15 NMR relaxation methods. J. Biomol. NMR 35, 225–239. (22) Li, J., Wu, H., Hong, J., Xu, X., Yang, H., Wu, B., Wang, Y., Zhu, J., Lai, R., Jiang, X., Lin, D., Prescott, M. C., and Rees, H. H. (2008) Odorranalectin Is a Small Peptide Lectin with Potential for Drug Delivery and Targeting. PLoS One 3. (23) Vakonakis, I., Langenhan, T., Promel, S., Russ, A., and Cambell, I. D. (2008) Solution structure and sugar-binding mechanism of mouse latrophilin-1 RBL: a 7TM receptor-attached lectin-like domain. Structure 16, 944–953. (24) Nesmelova, I. V, Ermakova, E., Daragan, V. A., Pang, M., Menéndez, M., Lagartera, L., Solís, D., Baum, L. G., and Mayo, K. H. (2010) Lactose Binding to Galectin-1 Modulates Structural Dynamics, Increases Conformational Entropy, and Occurs with Apparent Negative Cooperativity. J. Mol. Biol. 397, 1209–1230. (25) Diehl, C., Engström, O., Delaine, T., Håkansson, M., Genheden, S., Modig, K., Leffler, H., Ryde, U., Nilsson, U. J., and Akke, M. (2010) Protein Flexibility and Conformational Entropy in Ligand Design Targeting the Carbohydrate Recognition Domain of Galectin-3. J. Am. Chem. Soc. 132, 14577–14589. (26) Miller, M. C., Ribeiro, J. P., Roldos, V., Martin-Santamaria, S., Canada, F. J., Nesmelova, I. A., Andre, S., Pang, M., Klyosov, A. A., Baum, L. G., Jimenez-Barbero, J., Gabius, H. J., and Mayo, K. H. (2011) Structural aspects of binding of alpha-linked digalactosides to human galectin-1. Glycobiology 21, 1627–1641. (27) Ortega, G., Castaño, D., Diercks, T., and Millet, O. (2012) Carbohydrate Affinity for the Glucose–Galactose Binding Protein Is Regulated by Allosteric Domain Motions. J. Am. Chem. Soc. 134, 19869–19876. (28) Ermakova, E., Miller, M. C., Nesmelova, I. V, Lopez-Merino, L., Berbis, M. A., Nesmelov, Y., Tkachev, Y. V, Lagartera, L., Daragan, V. A., Andre, S., Canada, F. J., Jimenez-Barbero, J., Solis, D., Gabius, H. J., and Mayo, K. H. (2013) Lactose binding to human galectin-7 (p53-induced gene 1) induces long-range effects through the protein resulting in increased dimer stability and evidence for positive cooperativity. Glycobiology 23, 508–523. (29) Probert, F., Whittaker, S. B. M., Crispin, M., Mitchell, D. A., and Dixon, A. M. (2013) Solution NMR analyses of the C-type carbohydrate recognition domain of DC-SIGNR protein reveal different binding modes for HIV-derived oligosaccharides and smaller glycan fragments. J. Biol. Chem. 288, 22745–22757. (30) Canales, Á., Mallagaray, Á., Berbís, M. Á., Navarro-Vázquez, A., Domínguez, G.,

ACS Paragon Plus Environment

18

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Cañada, F. J., André, S., Gabius, H.-J., Pérez-Castells, J., and Jiménez-Barbero, J. (2014) Lanthanide-Chelating Carbohydrate Conjugates Are Useful Tools To Characterize Carbohydrate Conformation in Solution and Sensitive Sensors to Detect Carbohydrate– Protein Interactions. J. Am. Chem. Soc. 136, 8011–8017. (31) Rodriguez-Mias, R. A., and Pellecchia, M. (2003) Use of Selective Trp Side Chain Labeling To Characterize Protein−Protein and Protein−Ligand Interactions by NMR Spectroscopy. J. Am. Chem. Soc. 125, 2892–2893. (32) Bista, M., Kowalska, K., Janczyk, W., Dömling, A., and Holak, T. A. (2009) Robust NMR Screening for Lead Compounds Using Tryptophan-Containing Proteins. J. Am. Chem. Soc. 131, 7500–7501. (33) Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: A multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277–293. (34) Vranken, W. F., Boucher, W., Stevens, T. J., Fogh, R. H., Pajon, A., Llinas, P., Ulrich, E. L., Markley, J. L., Ionides, J., and Laue, E. D. (2005) The CCPN data model for NMR spectroscopy: Development of a software pipeline. Proteins-Structure Funct. Bioinforma. 59, 687–696. (35) Sattler, M., Schleucher, J., and Griesinger, C. (1999) Heteronuclear multidimensional NMR experiments for the structure determination of proteins in solution employing pulsed field gradients. Prog. Nucl. Magn. Reson. Spectrosc. 34, 93–158. (36) Yamazaki, T., Forman-Kay, J. D., and Kay, L. E. (1993) Two-dimensional NMR experiments for correlating carbon-13.beta. and proton.delta./.epsilon. chemical shifts of aromatic residues in 13C-labeled proteins via scalar couplings. J. Am. Chem. Soc. 115, 11054–11055. (37) Williamson, M. P. (2013) Using chemical shift perturbation to characterise ligand binding. Prog. Nucl. Magn. Reson. Spectrosc. 73, 1–16. (38) Farrow, N. A., Muhandiram, R., Singer, A. U., Pascal, S. M., Kay, C. M., Gish, G., Shoelson, S. E., Pawson, T., Formankay, J. D., and Kay, L. E. (1994) Backbone Dynamics of a Free and a Phosphopeptide-Complexed Src Homology 2 Domain Studied by 15N NMR Relaxation. Biochemistry 33, 5984–6003. (39) Rule, G. S., and Hitchens, T. K. (2006) Fundamentals of protein NMR Spectroscopy, in Focus on Structural Biology. Springer, Dordrecht, Netherlands. (40) Mandel, A. M., Akke, M., and Palmer III, A. G. (1995) Backbone Dynamics of Escherichia coli Ribonuclease HI: Correlations with Structure and Function in an Active

ACS Paragon Plus Environment

19

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

Enzyme. J. Mol. Biol. 246, 144–163. (41) Palmer III, A. G., Kroenke, C. D., and Loria, J. P. (2001) Nuclear magnetic resonance methods for quantifying microsecond-to-millisecond motions in biological macromolecules, in Methods in Enzymology (Thomas L. James, V. D., and Uli, S., Eds.), pp 204–238. Academic Press. (42) Bieri, M., and Gooley, P. (2011) Automated NMR relaxation dispersion data analysis using NESSY. BMC Bioinformatics 12, 421. (43) Farrow, N., Zhang, O., Forman-Kay, J., and Kay, L. (1994) A heteronuclear correlation experiment for simultaneous determination of 15N longitudinal decay and chemical exchange rates of systems in slow equilibrium. J. Biomol. NMR 4, 727–734. (44) Miloushev, V. Z., Bahna, F., Ciatto, C., Ahlsen, G., Honig, B., Shapiro, L., and Palmer III, A. G. (2008) Dynamic Properties of a Type II Cadherin Adhesive Domain: Implications for the Mechanism of Strand-Swapping of Classical Cadherins. Structure 16, 1195–1205. (45) Lis, H., and Sharon, N. (1998) Lectins: Carbohydrate-specific proteins that mediate cellular recognition. Chem. Rev. 98, 637–674. (46) Park, D., Ryu, K.-S., Choi, D., Kwak, J., and Park, C. (2007) Characterization and role of fucose mutarotase in mammalian cells. Glycobiology 17, 955–962. (47) Loris, R., Tielker, D., Jaeger, K.-E., and Wyns, L. (2003) Structural Basis of Carbohydrate Recognition by the Lectin LecB from Pseudomonas aeruginosa. J. Mol. Biol. 331, 861–870. (48) Berjanskii, M. V, and Wishart, D. S. (2005) A Simple Method To Predict Protein Flexibility Using Secondary Chemical Shifts. J. Am. Chem. Soc. 127, 14970–14971. (49) Jarymowycz, V. A., and Stone, M. J. (2006) Fast time scale dynamics of protein backbones: NMR relaxation methods, applications, and functional consequences. Chem. Rev. 106, 1624–1671. (50) Loria, J. P., Berlow, R. B., and Watt, E. D. (2008) Characterization of enzyme motions by solution NMR relaxation dispersion. Acc Chem Res 41, 214–221. (51) Kleckner, I. R., and Foster, M. P. (2011) An introduction to NMR-based approaches for measuring protein dynamics. Biochim. Biophys. Acta-Proteins Proteomics 1814, 942–968.

ACS Paragon Plus Environment

20

Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Supporting Information The supporting information may be accessed free of charge online at http://pubs.acs.org at DOI: Assigned 1H-15N HSQC spectra of RSL in the sugar-free and -bound forms. Plots of chemical shift perturbations due to sugar binding. 1H-15N HSQC spectra showing α- and β-anomer effects. Structural representation of RSL showing the R2ex data.

ACS Paragon Plus Environment

21

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 29

Figures

Figure 1. (A) The primary structure of RSL comprises an N- and C-terminal half, which share ~40 % sequence identity. Colons and dots indicate identical and similar residues, respectively, as revealed by (B) a structural alignment. Residues belonging to the intra- and intermonomeric binding sites are shown in grey and black, respectively. The panel on the right shows how the methyl substituent of the sugar is exposed to the solvent. Side chains shown as sticks (except Thr82) make at least one non-covalent bond with the sugar.11

ACS Paragon Plus Environment

22

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 2. Spectral regions from the overlaid 1H-15N HSQC spectra of sugar-free RSL (black contours) and RSL bound to L-fucose (red), to methyl-α-L-fucoside (navy) or methyl-β-Lfucoside (orange). (A) An example of structurally homologous residues, R17 and R62, with resonances of similar chemical shift. (B) Example of a resonance, A85, that is sensitive to both anomers of L-fucose. (C) The seven tryptophan indoles are useful reporters of sugar binding (refer to Figure 4).

ACS Paragon Plus Environment

23

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 29

Figure 3. ∆δavg plots of RSL in the presence of 2 equivalents of different monosaccharides. Black bars indicate resonances that were split in two due to the presence of the α- and βanomer. Blanks correspond to prolines (P14 and P44) and residues which lacked a resonance in the sugar-free or -bound RSL. The structure of each sugar and their Kd values11,19 are indicated. D-mannose is oriented to show how hydroxyls O2, O3 and O4 match the stereochemistry of L-fucose.

ACS Paragon Plus Environment

24

Page 25 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 4. (A) Plot of chemical shift perturbations for the tryptophan indole resonance in the presence of four monosaccharides (refer to Figure 3). Filled and open circles correspond to the 1HN and 15N data, respectively. Some resonances were split in two (e.g. L-fucose data) due to the presence of the α- and β-anomer. (B) A simplified representation of the binding site showing the homologous pairs (refer to Figure 1) of tryptophan side chains as sticks and colour-coded to match panel A. The methyl-α-L-fucoside is shown as lines and the protein backbone is represented as a ribbon. Two crystallographic water molecules are indicated by red spheres. Dashed lines indicate the distances between the Trp Nε atoms and C6 or O3 of methyl-α-L-fucoside.

ACS Paragon Plus Environment

25

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 29

Figure 5. Regions from the overlaid 1H-15N HSQC spectra of sugar-free RSL (black contours) and RSL bound to methyl-α-L-fucoside (navy), methyl-β-L-fucoside (dark orange), nitrophenyl-α-L-fucoside (light blue) or nitrophenyl-β-L-fucoside (light orange). The similarities / dissimilarities in chemical shift perturbations due to sugar binding are evident. β substituents resulted in a distinct pattern (see C75, T82 and G84).

ACS Paragon Plus Environment

26

Page 27 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 6. Squared order parameters (S2), fast dynamics (15N R2/R1 profiles), and exchange contribution to the transverse relaxation rates (R2ex) of RSL in the sugar-free and -bound forms. S2 values were calculated in RCI48 from the backbone amide secondary chemical shifts. Note the flexible (S2 < 0.75) loop regions comprising residues 32-35 and 77-80. The dashed lines are the averages for the sugar-free data. The RSL secondary structure is indicated.

ACS Paragon Plus Environment

27

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 29

Figure 7. Slow dynamics of RSL bound to L-fucose. Selected spectral regions of the 15N ZZexchange experiments acquired with tmix = 0 s (black), 0.23 s (blue), and 0.5 s (red) showing the resonances of (A) W81 and W36 (indoles) and (B) A85. Dashed rectangles connect the four exchange cross-peaks. Note the intensity decrease of the auto-peaks (AA, BB) and the concomitant increase of the cross-peaks (AB, BA) at longer mixing times. (C) Quantitative analysis of the build-up curve for the exchange cross-peaks of A85. See Experimental Section for the fit parameters.

ACS Paragon Plus Environment

28

Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Table of Contents Graphic

ACS Paragon Plus Environment

29