Insights into substrate specificity of archaeal Entner ... - ACS Publications

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Insights into substrate specificity of archaeal Entner-Doudoroff aldolases: the structures of Picrophilus torridus 2-keto-3-deoxygluconate aldolase, and Sulfolobus solfataricus 2-keto-3-deoxy(6-phospho)gluconate aldolase in complex with 2-keto-3-deoxy-6-phosphogluconate. Viatcheslav Zaitsev, Ulrike Johnsen, Matthias Reher, Marius Ortjohann, Garry L. Taylor, Michael J Danson, Peter Schönheit, and Susan Crennell Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b00535 • Publication Date (Web): 29 May 2018 Downloaded from http://pubs.acs.org on May 29, 2018

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 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 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.

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 41 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

1

Insights into substrate specificity of archaeal Entner-Doudoroff aldolases: the structures of

2

Picrophilus torridus 2-keto-3-deoxygluconate aldolase, and Sulfolobus solfataricus 2-keto-3-

3

deoxy(6-phospho)-gluconate aldolase in complex with 2-keto-3-deoxy-6-phosphogluconate.

4

Viatcheslav Zaitseva, Ulrike Johnsenb, Matthias Reherb, Marius Ortjohannb, Garry L.Taylora, Michael J.

5

Dansonc, Peter Schönheitb and Susan J. Crennellc#

6 7

Biomolecular Sciences, University of St Andrews, St Andrews, Fife, UKa; Institut für Allgemeine

8

Mikrobiologie, Christian-Albrechts-Universität, Kiel, Germanyb; Department of Biology & Biochemistry,

9

University of Bath, Bath, UKc.

10 11

V.Z. and U.J. contributed equally to this work

12 13

# Corresponding author: Susan Crennell, [email protected]

14 15

Second contributing author: Peter Schönheit [email protected]

16 17

Running title: Picrophilus and Sulfolobus KDG-Aldolase specificity

18 19

Abbreviations.

20 21

KDG:

2-keto-3-deoxygluconate

22

KDPG:

2-keto-3-deoxy-6-phosphogluconate

23

KDG-aldolase:

2-keto-3-deoxygluconate aldolase

24

KDGK:

KDG kinase

25

KD(P)G-aldolase:

2-keto-3-deoxygluconate/2-keto-3-deoxy-6-phosphogluconate aldolase

26

Pt-KDG-aldolase:

Picrophilus torridus KDG-aldolase

27

RMSD:

Root Mean Square Deviation

1 ACS Paragon Plus Environment

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

28

Ss-KD(P)G-aldolase: Sulfolobus solfataricus KD(P)G-aldolase

29 30

Table of Contents graphic (Figure7)

31 32

EC number: 4.1.2.51

33 34

Keywords:

Picrophilus torridus, 2-keto-3-deoxygluconate aldolase, crystal structure, enzyme

35

specificity, Sulfolobus solfataricus, 2-keto-3-deoxy-phospho-gluconate aldolase,

36

archaeal Entner-Doudoroff pathways

37


Page 2 of 41

Page 3 of 41 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

38

Biochemistry

Abstract

39 40

The thermoacidophilic archaea Picrophilus torridus and Sulfolobus solfataricus catabolise glucose via

41

a non-phosphorylative Entner-Doudoroff pathway and a branched Entner-Doudoroff pathway,

42

respectively. Key enzymes for these Entner-Doudoroff pathways are the aldolases, 2-keto-3-

43

deoxygluconate aldolase (KDG-aldolase) and 2-keto-3-deoxy-6-phosphogluconate aldolase (KD(P)G-

44

aldolase). KDG-aldolase from P. torridus (Pt-KDG-aldolase) is highly specific for the non-

45

phosphorylated substrate, 2-keto-3-deoxygluconate (KDG), whereas KD(P)G-aldolase from

46

S. solfataricus (Ss-KD(P)G-aldolase) is an enzyme catalyzing the cleavage of both KDG and 2-keto-3-

47

deoxy-6-phosphogluconate (KDPG), with a preference for KDPG. The structural basis for the high

48

specificity of Pt-KDG-aldolase for KDG as compared to the more promiscuous Ss-KD(P)G-aldolase

49

has not been analysed before. In the current paper we report the elucidation of the structure of Ss-

50

KD(P)G-aldolase in complex with KDPG at 2.35Å and that of KDG-aldolase from P. torridus at 2.50 Å

51

resolution. By superimposition of the active sites of the two enzymes, and subsequent site-directed

52

mutagenesis studies, a network of four amino acids, namely Arg106, Tyr132, Arg237 and Ser241, was

53

identified in Ss-KD(P)G-aldolase that interact with the negatively-charged phosphate group of KDPG,

54

thereby raising the affinity of the enzyme for KDPG. This KDPG-binding network is absent in Pt-KDG-

55

aldolase, which explains the low catalytic efficiency of KDPG cleavage.

56 57


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 41

58

Introduction

59

The discovery of thermophilic micro-organisms has stimulated a wealth of fundamental and applied

60

research into their enzymes, both for the light they may shed on the molecular basis of protein stability

61

and for the potential applications of enzymes with high thermal stability. Most extreme thermophiles

62

are members of the Archaea, the third domain of life, and studies on their metabolic pathways have

63

revealed enzymes with unusual catalytic activities and specificities.

64

The Archaea degrade glucose and glucose polymers to pyruvate, not via the classical

65

Embden-Meyerhof pathway nor the Entner-Doudoroff pathway reported for Bacteria and Eukarya, but

66

via modifications of these pathways, thereby implicating novel enzymes and enzyme families with

67

unusual catalytic and regulatory features.1,2,3 Modified Embden-Meyerhof pathways have been

68

described mainly in anaerobic hyperthermophilic archaea whereas aerobic archaea, including the

69

thermoacidophiles Picrophilus torridus and Sulfolobus solfataricus, degrade glucose via modified

70

versions of the Entner-Doudoroff pathway. For P. torridus, growing optimally at 60°C and pH 0.9, a

71

non-phosphorylative Entner-Doudoroff pathway was reported.4,5 Accordingly, glucose is converted to

72

2-keto-3-deoxygluconate (KDG) via glucose dehydrogenase and gluconate dehydratase. KDG is then

73

cleaved by a KDG-aldolase to pyruvate and glyceraldehyde, of which the latter is converted to a

74

second molecule of pyruvate via glyceraldehyde dehydrogenase, 2-phosphoglycerate-forming

75

glycerate kinase, enolase and pyruvate kinase (Fig. 1).

76


Page 5 of 41 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

77 Non-phosphorylative ED pathway P. torridus

Branched ED pathway S. solfataricus

Glucose GDH

Glucose GDH

NAD(P)H

Gluconate

NAD(P)H

Gluconate

GAD

KDGK

GAD

ATP

KDG

KDPG

KDG

KDGA

KD(P)GA

Pyruvate GA

Pyruvate GA

GADH

GAOR

NAD(P)H

Glycerate

XH2

Glycerate

ATP

GLK

GAP Pyruvate

NAD(P)H

ATP

GLK

2-PG

GAPN

3-PG

2-PG PGM

ENO

ENO PEP

PYK

PEP ADP

PYK

Pyruvate

ADP

Pyruvate

78 79

Fig 1 Modified Entner-Doudoroff (ED) pathways in P. torridus and S. solfataricus. KDG, 2-keto-3-

80

deoxygluconate; GA, glyceraldehyde; 2-PG, 2-phosphoglycerate; PEP, phosphoenolpyruvate; KDPG,

81

2-keto-3-deoxy-6-phosphogluconate; GAP, glyceraldehyde-3-phosphate; 3-PG, 3-phosphoglycerate;

82

GDH, glucose dehydrogenase; GAD, gluconate dehydratase; KDGA, KDG aldolase; GADH,

83

glyceraldehyde dehydrogenase; GLK, glycerate kinase (2-PG forming); ENO, enolase; PYK, pyruvate

84

kinase; KD(P)GA, KDG/KDPG aldolase; KDGK, KDG kinase; GAOR, glyceraldehyde oxidoreductase;

85

GAPN, non-phosphorylative GAP dehydrogenase; PGM, phosphoglycerate mutase. The C4 epimeric

86

substrate galactose and its respective derivatives of GDH, GAD, KDGK, KDGA and KD(P)GA are not

87

shown.


Biochemistry 1 2 88 3 4 89 5 6 90 7 8 91 9 10 11 92 12 13 93 14 15 94 16 17 95 18 19 96 20 21 97 22 23 98 24 25 99 26 27 100 28 29 101 30 31 32 102 33 34 103 35 36 104 37 38 105 39 40 106 41 42 107 43 44 108 45 46 109 47 48 110 49 50 51 111 52 53 112 54 55 113 56 57 58 59 60

Page 6 of 41

On the other hand, S. solfataricus, growing at 80°C and pH 3.5, possesses a branched EntnerDoudoroff pathway.6,7 In this pathway, glucose is converted to KDG via glucose dehydrogenase and gluconate dehydratase as in P. torridus. Further catabolism of KDG involves two branches; in the first, KDG is cleaved to pyruvate and glyceraldehyde, which is further converted to pyruvate via glyceraldehyde oxidoreductase, glycerate kinase, enolase and pyruvate kinase. In the second branch, KDG is phosphorylated by KDG kinase to generate 2-keto-3-deoxy-6-phosphogluconate (KDPG), which is cleaved to pyruvate and glyceraldehyde-3-phosphate by a KD(P)G-aldolase that catalyzes the cleavage of both KDG and KDPG. Further conversion of glyceraldehyde-3-phosphate to pyruvate involves a non-phosphorylative glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate mutase, enolase, and pyruvate kinase (Fig. 1). Both the non-phosphorylative Entner-Doudoroff pathway of P. torridus and the branched Entner-Doudoroff pathway of S. solfataricus have been described to be promiscuous for the catabolism of both D-glucose and D-galactose .4,5,7,8 The substrate specificities of the Picrophilus and Sulfolobus aldolases are key determinants of their respective metabolic pathways. KDG-aldolase (E.C. 4.1.2.51) from P. torridus is highly specific for the substrate KDG, showing up to 2000-fold higher catalytic efficiency compared to KDPG.4 In contrast, KD(P)G-aldolases are enzymes catalyzing the cleavage of both KDG and KDPG, with a preference for KDPG. KD(P)G-aldolases have been characterized from the Sulfolobus species, S. solfataricus, S. acidocaldarius and S. tokodaii, and from Thermoproteus tenax .6,7,9 Both types of aldolases, KDG-aldolase and KD(P)G-aldolase, are Schiff-base-forming class I aldolases that belong to the dihydrodipicolinate synthase-like family of the class I aldolase superfamily, where they form distinct clusters in accordance with their different substrate specificities .4 It should be noted that the archaeal KDG-aldolase and KD(P)G-aldolase differ from the KDPG aldolases of the classical EntnerDoudoroff pathway of bacteria and of the semi-phosphorylative Entner-Doudoroff pathway of the haloarchaeon Haloferax volcanii,10 which belong to the bacterial type class I KDPG aldolases/4-


Page 7 of 41 1 2 114 3 4 115 5 6 116 7 8 117 9 10 118 11 12 119 13 14 15 120 16 17 121 18 19 122 20 21 123 22 23 124 24 25 125 26 27 126 28 29 127 30 31 128 32 33 34 129 35 36 130 37 38 131 39 40 132 41 42 133 43 44 134 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

hydroxy-2-oxoglutarate aldolase family .11 While the differences in substrate specificities of KDG-aldolase and KD(P)G-aldolase are in accordance with their in vivo operation in the non-phosphorylative and branched Entner-Doudoroff pathways, respectively, the structural basis for the high specificity of Pt-KDG-aldolase for KDG as compared to the more promiscuous KD(P)G-aldolase from Sulfolobus has not been analysed so far. However,

the

structures

of

KD(P)G-aldolases

from

S.

solfataricus

(Ss-KD(P)G-aldolase),

S. acidocaldarius and T. tenax have been determined. KDPG was modelled into the active site of the S. acidocaldarius KD(P)G-aldolase

9

and of T. tenax KD(P)G-aldolase,12 and from these structural

models two arginines and one tyrosine have been proposed to contact the phosphate group of KDPG. To analyse the different substrate specificities of the aldolases further, a direct structural comparison of the KDG-aldolase with KD(P)G-aldolase is necessary. Therefore, in the current paper, we report the elucidation of the structure of KDG-aldolase from P. torridus, specific for the nonphosphorylated substrate KDG, and also the structure of the KD(P)G-aldolase from S. solfataricus in complex with KDPG. By subsequent superimposition of the active sites of the two enzymes, we identify amino acids that contact the phosphate group of KDPG in Ss-KD(P)G-aldolase and the corresponding residues in Pt-KDG-aldolase. Subsequent predictions of the structural basis of the respective substrate specificities are then tested by site-directed mutagenesis.


Biochemistry 1 2 135 3 4 5 136 6 7 137 8 9 138 10 11 139 12 13 140 14 15 16 141 17 18 142 19 20 143 21 22 144 23 24 145 25 26 146 27 28 147 29 30 148 31 32 149 33 34 35 150 36 37 151 38 39 152 40 41 153 42 43 154 44 45 155 46 47 156 48 49 157 50 51 158 52 53 54 159 55 56 160 57 58 59 60

Page 8 of 41

Materials and Methods Site-directed mutagenesis Variants of KDG aldolase from P. torridus (Pt-KDG-aldolase) and of KD(P)G aldolase from S. solfataricus (Ss-KD(P)G-aldolase) were generated using the QuikChange Lightning Site-directed Mutagenesis Kit (Agilent Technologies, Waldbronn, Germany) or the Q5 Site-directed Mutagenesis Kit (New England Biolabs, Frankfurt, Germany), according to the manufacturers’ instructions. The plasmids pET3a-SSO_3197 8 and pET19b-Pto1279a 4 were used as templates.

Expression and purification of enzymes Pt-KDG-aldolase and Ss-KD(P)G-aldolase, and their respective variants, were expressed in E. coli (DE3) Rosetta pLysS cells grown at 37°C for 4 h in Lysogeny Broth (LB) medium. Expressions were induced by the addition of isopropyl β-D-1-thiogalactopyranoside. Recombinant Pt-KDG-aldolase and variants of the enzyme were purified by heat treatment at 55°C for 30 min, followed by affinity chromatography on Ni-NTA Agarose according to Reher et al. (2010). Eluted KDG-aldolase was then applied to a Superdex 200 HiLoad 16/60 column (GE Healthcare, Freiburg, Germany) that was equilibrated with 50 mM Tris-HCl, pH 7.4, containing 150 mM NaCl. Protein was eluted with an isocratic flow. At this stage, Pt-KDG-aldolase and variants were essentially pure as judged by sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Cell pellets of recombinant Ss-KD(P)G-aldolase and site-directed variants were suspended in 50 mM Tris-HCl, pH 8.6, and passed three times through a French press cell; cell extracts were generated by centrifugation at 39,000 g. Purification was performed by heat treatment at 75°C for 30 min followed by centrifugation at 39,000 g. The resulting supernatant was applied to a Q Sepharose HP HiLoad 16/10 column (GE Healthcare) that was equilibrated in 50 mM Tris-HCl, pH 8.6. The protein was eluted with an increasing gradient from 0 to 2 M NaCl. Fractions containing highest KD(P)G-aldolase activity were applied to a Superdex 200 HiLoad 16/60 column. Elution of protein was performed in 50 mM Tris-HCl, pH 7.4, containing 150 mM NaCl. After this procedure, the protein was


Page 9 of 41 1 2 161 3 4 162 5 6 163 7 8 164 9 10 165 11 12 166 13 14 15 167 16 17 168 18 19 169 20 21 170 22 23 171 24 25 172 26 27 173 28 29 174 30 31 175 32 33 34 176 35 36 177 37 38 178 39 40 179 41 42 180 43 44 181 45 46 182 47 48 183 49 50 51 184 52 53 185 54 55 186 56 57 58 59 60

Biochemistry

essentially pure as judged by SDS-PAGE.

Determination of enzyme activity Aldol-cleavage activities of Pt-KDG-aldolase and Ss-KD(P)G-aldolase were measured by following the production of pyruvate from KDG or KDPG. Assays were carried out at 60°C in 50 mM sodium phosphate buffer, pH 6.2, containing 0.3 mM NADH, 3 U lactate dehydrogenase, and KDPG (SigmaAldrich, Schnelldorf, Germany; product 79156) or KDG (Sigma-Aldrich, product 12271); the reduction of

pyruvate to

lactate

by

NADH,

catalysed

by

lactate dehydrogenase,

was

monitored

spectrophotometrically at 340nm.4

Crystallisation, X-ray data collection, processing and model refinement The genes encoding Pt-KDG-aldolase and Ss-KD(P)G-aldolase were cloned and expressed, and the recombinant enzymes purified and characterised, as described above. For crystallization, Pt-KDGaldolase was concentrated to 4.2 mg/ml and, following the identification of initial crystallisation conditions (Crystal Screen Cryo HT, Hampton Research), optimal crystals were obtained by the hanging-drop, vapour-diffusion method, using drops of protein and well solution (0.085 M sodium HEPES buffer pH 7.5, 8.5% (v/v) 2-propanol, 17% (w/v) PEG 4000, 15% (v/v) anhydrous glycerol) in equal ratio. Ss-KD(P)G-aldolase, concentrated to 6mg/ml, was also crystallised by the hanging-drop, vapour-diffusion method using drops containing 1.5 µl of protein mixed with 1.5 µl of well solution (0.1M sodium HEPES buffer pH 6.0, 8% (v/v) 2-propanol, 11% (w/v) PEG4000)13. The Ss-KD(P)Galdolase crystals were soaked in well solution containing 15mM KDPG (Sigma) and 20% v/v glycerol (as cryoprotectant) for 5 min, followed by flash-cooling in liquid nitrogen. X-ray data were collected from Pt-KDG-aldolase crystals at 100K using a Rigaku MicroMax 007 HF with Saturn 944+ CCD detector. Diffraction spots were visible to at least 2.4Å; however, during data processing the resolution was truncated to 2.5Å to optimise data completeness. X-ray data were processed and scaled using HKL-2000.14 The molecular replacement pipeline BALBES15 was used to


Biochemistry 1 2 187 3 4 188 5 6 189 7 8 190 9 10 191 11 12 192 13 14 15 193 16 17 194 18 19 195 20 21 196 22 23 197 24 25 198 26 27 199 28 29 200 30 31 32 201 33 34 202 35 36 203 37 38 204 39 40 205 41 42 206 43 44 207 45 46 208 47 48 209 49 50 51 210 52 211 53 54 55 56 57 58 59 60

Page 10 of 41

find initial phases using the structure of Bacillus anthracis 4-hydroxy-tetrahydrodipicolinate synthase DapA-2 (1XKY,16 with which Pt-KDG-aldolase shares 32% identity) as the search model. Iterative cycles of model building in Coot17 and refinement in Phenix 18 were required to build a final model. X-ray data were collected from the SsKD(P)G-aldolase complex crystals at 100K on the I04 beamline at the Diamond Light source, using a Dectris Pilatus 6M-F detector. 720° of data were collected using radiation of wavelength 0.9795Å and processed using DIALS,19 then scaled using AIMLESS,20 truncated to 2.35Å to achieve completeness, with high CC(1/2) and sufficient signal (mean I/σ(I)). BALBES15 was again used to solve the structure, using the unliganded Ss-KD(P)Galdolase structure (1W37) as the starting model, and cycles of rebuilding in Coot and refinement in Phenix. The KDPG ligand structure was built and the Schiff base conformation and CIF file produced using JLigand.21 Except where indicated, PyMOL22 was used to produce the figures in this paper.

Size-exclusion chromatography For determination of the native relative molecular masses (Mr) of Pt-KDG-aldolase, Ss-KD(P)Galdolase and their respective variants, a Superdex 200 HiLoad 16/60 column was calibrated with highmolecular-weight (HMW) and low-molecular-weight (LMW) kits (GE Healthcare, Germany) as specified by the manufacturer. The subunit Mr values were calculated from the protein sequences and were seen to be consistent with estimates from Coomassie-stained SDS-PAGE.

Accession number: The Pt-KDG-aldolase structure has been deposited in the PDB and given the accession number 4UXD, DOI: 10.2210/pdb4uxd/pdb. The Ss-KD(P)G-aldolase structure in complex with KDPG has been given the accession number 6G3Z.


Page 11 of 41 1 2 212 3 4 213 5 214 6 7 215 8 9 216 10 11 217 12 13 14 218 15 16 219 17 18 220 19 20 221 21 22 222 23 24 223 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 52 224 53 54 225 55 226 56 227 57 58 59 60

Biochemistry

Results and Discussion

Overall Pt-KDG-aldolase structure: The structure of Pt-KDG-aldolase has been determined to 2.5 Å, deposited in the PDB and given the code 4UXD. Data processing and structure refinement data are given in Table 1. The asymmetric unit contains one Pt-KDG-aldolase tetramer, 19 glycerol (cryoprotectant) molecules, three molecules of propan-2-ol, five of ethane-1,2-diol, two PEG derivatives (PEG, PGE), and 340 water molecules. The structure is of high quality, with a good Molprobity score (1.90, 97th percentile), although 6 residues were classified as Ramachandran plot outliers, these being residue 105 in chains A, C and D, residue 77 in chains A and C, and residue 207 in chain C. All these residues are marginal outliers, not seen in all chains; however, the most consistent outlier is Tyr105 and its unusual conformation can be seen to be caused by its position on the dimer interface, projecting into a pocket in the other monomer (Fig. 2).

Fig 2: The structure of Pt-KDG-aldolase in cartoon representation, with each monomer in a different colour. The Ramachandran-outliers Tyr 105 are shown in stick form, and can be seen to protrude into a neighbouring molecule as part of the larger dimer interface, the perpendicular tetramer interface being less extensive.


Biochemistry 1 2 228 3 4 229 5 6 230 7 8 231 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 12 of 41

Table 1: Data processing and model refinement statistics. Numbers in parentheses refer to the highest resolution shell in each data set (2.54-2.50Å for Pt-KDG-aldolase, 2.43-2.35Å for Ss-KD(P)Galdolase).

Data processing

Pt-KDG-aldolase

Ss-KD(P)G-aldolase

Unit Cell Dimensions (Å)

77.9, 100.2, 154.6

106.9, 106.9, 245.6

Space Group

P22121

P6522

Resolution (Å)

28.58-2.50 (2.54-2.50)

92.55-2.35 (2.43-2.35)

Completeness (%)

86.6 (83.4)

100.0 (100.0)

Rmerge

0.107 (0.333)

0.145 (2.51)

Rpim (all I+ & I-)

0.017(0.284)

Redundancy

3.8 (2.8)

75.8 (79.1)

I/σI

13.2 (2.7)

22.2 (2.7)

Number of observed reflections

877347

2692983 (269399)

Number of unique reflections

36955 (1730)

35510 (3406)

Number of working set reflections

34773

35153

Number of test set reflections

1884

1686

Final Rcryst

0.175

0.187

Final Rfree

0.260

0.239

Number of non-H protein atoms

8979

4838

Number of non-H solvent atoms

340

79

Model refinement


Page 13 of 41 1 2 3 4 5 6 7 8 9 10 11 12 232 13 14 233 15 16 234 17 18 19 235 20 21 236 22 23 237 24 25 238 26 27 239 28 29 240 30 31 241 32 33 242 34 35 243 36 37 38 244 39 40 245 41 42 246 43 44 247 45 46 248 47 48 249 49 50 250 51 52 53 251 54 55 252 56 57 253 58 59 60

Biochemistry

RMSD bonds (Å)

0.008

0.009

RMSD angles (°)

1.18

1.00

Ramachandran plot most favoured (%)

95.8

96.74

Ramachandran plot allowed (%)

3.7

3.26

The asymmetric unit contains 4 chains, each of which folds into a classic (βα)8 TIM-barrel structure, and their structural similarity is shown by the root mean square deviation (RMSD) values of any pair of superimposed independently-refined monomers being no greater than 0.3Å. Analysis of PtKDG-aldolase by PISA23 indicates that these monomers form a tetramer as a dimer of dimers, with the average dimer interface having an area of 1610Å2 and a solvation energy for folding (∆Gf) of -20 kcal/mol, while that between dimers in the tetramer is only 680Å2 and -5.6kcal/mol. Comparison with other structures in the PDB using PISA clearly identifies Pt-KDG-aldolase as a member of the dihydrodipicolinate synthetase family, having marked structural similarity with the many structures in 'Aldolase Class 1' (CATH superfamily 3.20.20.70). Several structures of KD(P)Galdolase have been published, including those from T. tenax (RMSD to Pt-KDG-aldolase of 1.9Å over 259 Cα, sequence identity 21.7%), Sulfolobus acidocaldarius (2.0Å over 274 Cα, sequence identity 27.9%), and S. solfataricus (1.8Å over 259 Cα, sequence identity 24.8%). Despite the low sequence identity between these KD(P)G-aldolases, and the somewhat larger RMSD values, indicating less similarity with Pt-KDG-aldolase than the approximately 1.0Å RMSD calculated when these structures are compared to each other, the structural similarity is clear (see Fig. 3 for alignment of Ss-KD(P)GA and Pt-KDG-aldolase). All these structures have also been solved in the presence of pyruvate, glyceraldehyde or KDG, and no significant movement of the framework is seen on binding.

Overall structure of Ss-KD(P)G-aldolase in complex with KDPG. To investigate the interactions of KDPG with a KD(P)G-aldolase, the structure of Ss-KD(P)G-aldolase in complex with KDPG was determined to 2.35Å; data processing and model refinement statistics are


Biochemistry 1 2 254 3 4 255 5 6 256 7 8 257 9 10 11 258 12 13 259 14 15 260 16 17 261 18 19 262 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 50 263 51 264 52 53 265 54 55 266 56 57 58 59 60

Page 14 of 41

given in Table 1. The overall structure is very similar to both the unliganded (1W37) and the KDGbound (1W3N) structures (average RMSD over all 293 Cα: 0.3Å).

Following refinement of the

unliganded coordinates, there was clear difference density for a ligand bound in the active site, extending beyond the KDG seen in the 1W3N structure. (Fig 4B). However the density was too broad to fit a single phosphate position, and therefore the final model has KDPG in a dual conformation, with the two positions occupied approximately equally (0.52:0.48), and has been deposited in the PDB with the code 6G3Z.

Fig 3. Cartoon representation of Ss-KD(P)G-aldolase (grey) aligned to Pt-KDG-aldolase (green) to show the overall structural similarity. The active site is marked by a stick representation of the Lysine (155 and 159 respectively) which forms the Schiff base in the catalytic reaction, and which is bound to


Page 15 of 41 1 2 267 3 4 268 5 6 269 7 8 270 9 10 11 271 12 13 272 14 15 16 273 17 18 274 19 20 275 21 22 276 23 24 277 25 26 278 27 28 279 29 30 280 31 32 33 281 34 35 282 36 37 283 38 39 284 40 41 285 42 43 286 44 45 287 46 47 288 48 49 50 289 51 52 290 53 54 291 55 56 292 57 58 59 60

Biochemistry

KDPG (yellow) in the Ss-KD(P)G-aldolase complex structure. The three N-terminal amino acids of the Pt-KDG-aldolase are removed for clarity.

Active site The structure of Ss-KD(P)G-aldolase in complex with KDG has been described previously13. As noted by Theodossis et al. for this structure, in comparison with the unliganded enzyme (1W37), the majority of the Ss-KD(P)G-aldolase active site is unchanged on binding KDG, although there are minor movements of aromatic side chains (Phe 39 and Tyr 130) that are mirrored in the KDPG-bound structure. As well as these hydrophobic interactions, the same hydrogen-bonding interactions between the KDG moiety and the enzyme are seen in the two structures (the C1 acid group with the main chain of both Thr43 and Thr44 and the Thr44 side chain, O4 with Tyr130 and O6 with Tyr132 ) (Fig 4B). However, the KDPG-bound structure shows that the more distal part of the active site involved in phosphate interactions is somewhat less rigid; on binding KDPG, the Cα positions of Arg106, and in some monomers Glu159 (which forms salt bridges with Arg106 and Arg273; Fig 5B), are the only nonterminal Cα amino acids to move more than 1Å from their positions in the earlier structures. The movement of Arg106 can be explained through both phosphate conformations of KDPG making strong interactions (within hydrogen-bonding distance) with Arg106 from a neighbouring monomer (Fig 5B). The other interactions with the phosphates are from residues within the original monomer, conformer A interacting with Ser241 and Tyr132, while conformer B interacts with Arg237. Through modeling studies, the interactions with Arg237, Tyr132 and Arg106 have been predicted previously in Sa-KD(P)G-aldolase (234, 105, 131)9 and Tt-KD(P)G-aldolase (255, 150, 124)12, being conserved across other KD(P)G aldolases, while the Ser 241, which is not conserved, had not been identified in those studies. In all the KD(P)G-aldolase structures, the active site containing the lysine that will form the Schiff base is in a conserved position at the top of the β-barrel. As was also seen in S. acidocaldarius KD(P)G-aldolase when crystallised at pH 7.5, in the Pt-KDG-aldolase structure the lysine points away


Biochemistry

A

Page 16 of 41

B 1 2 293 3 4 294 5 6 295 7 8 296 9 10 11 297 12 13 298 14 15 299 16 300 17 18 19 301 20 21 302 22 23 303 24 25 304 26 27 305 28 29 306 30 31 307 32 33 34 308 35 36 37 309 38 39 310 40 41 311 42 43 312 44 45 313 46 47 314 48 49 315 50 51 316 52 53 317 54 55 56 318 57 58 59 60

from the pyruvate binding pocket into the centre of the barrel (Fig. 3); however, in S. acidocaldarius KD(P)G-aldolase it was shown that no other amino acids moved significantly between this arrangement and the catalytically-competent configuration seen at pH 6.3.9 The heightened mobility of this residue in Pt-KDG-aldolase is shown by the large temperature factor of the lysine Nζ (50.2Å2) relative to the average of side chain atoms (36.8Å2), and the reduced electron density at 1.5 σ(Figure 4A).

Fig 4. Following superimposition of the structures, the active sites of (A) the apo structure of Pt-KDGaldolase (GOL are glycerol molecules) and (B) Ss-KD(P)G-aldolase with KDPG (yellow bonds) bound are shown in the same orientation with the 2FoFc electron density map for each structure, contoured at 1.5σ, restricted to that around KDPG in B for clarity. B also contains the structure of Ss-KD(P)Galdolase with KDG bound (grey) superimposed to show the relative stasis of all but those amino acids that interact with the phosphates. The main chain cartoon containing Arg106 from a neighbouring monomer is omitted as it crosses the middle of the image. Hydrogen bonds are shown as dotted lines.

Comparison of the inner part of the binding-pocket seen in the Ss-KD(P)G-aldolase-KDPG complex structure ( Fig. 4B)20 with the Pt-KDG-aldolase active site (Fig. 4A) shows that many of the KDG interactions would be conserved in the Picrophilus enzyme; these comprise interactions of the pyruvate

carboxylate

group

with

Ss-KD(P)GA Thr43 (Pt-KDG-aldolase

Ser44), Thr44(45),

Tyr130(131), and the glyceraldehyde with Thr157(Ser161) and, via water, with Gly179(184), Thr44(45) and Ala198(Gly203). The suitability of the site to bind polar molecules in Pt-KDG-aldolase is shown by the presence of glycerol (cryoprotectant) molecules, making many of the same interactions as KDG in Ss-KD(P)G-aldolase (Fig. 4A). The interaction of the O6 of KDG with Ss-KD(P)G-aldolase Tyr132 would not be replicated in Pt-KDG-aldolase as the corresponding residue is Ile133. Both enzymes have shown promiscuity in catalyzing the cleavage of both KDG and KDGal,

4,7

so it is likely that the

mechanism of action, as observed in Ss-KD(P)G-aldolase,7 is conserved in Pt-KDG-aldolase due to


Page 17 of 41 1 2 319 3 4 320 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

the similarities in the inner part of the binding pocket.


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 33 321 34 35 322 36 37 323 38 39 324 40 41 325 42 43 326 44 45 327 46 47 328 48 49 329 50 51 330 52 53 331 54 55 56 332 57 58 59 60

A

Page 18 of 41

B

C

D

Fig 5: Analysis of interactions (shown in magenta dotted lines) made by KDPG with the active sites of Pt-KDG-aldolase (modeled on the complex between KDPG and Ss-KD(P)G-aldolase) (A,C), and those seen in Ss-KD(P)G-aldolase (B,D). (A) Stick representations of amino acids from Pt-KDGaldolase (green) that could interact with KDPG (magenta), with those amino acids from neighbouring subunits in different shades of green. (B) A similar representation of amino acids in the Ss-KD(P)Galdolase active site, with those from a neighbouring monomer in a different shade of blue. (C) Electrostatic surface around the catalytic site of Pt-KDG-aldolase, with positively-charged areas in blue and negative in red. (D) Electrostatic surface around the catalytic site of Ss-KD(P)G-aldolase, coloured similarly.


Page 19 of 41 1 2 333 3 4 334 5 6 335 7 8 336 9 10 337 11 12 338 13 14 15 339 16 17 340 18 19 341 20 21 342 22 23 343 24 25 344 26 27 345 28 29 346 30 31 347 32 33 34 348 35 36 349 37 38 350 39 40 351 41 42 352 43 44 353 45 46 354 47 48 355 49 50 356 51 52 53 357 54 55 358 56 57 359 58 59 60

Biochemistry

Unlike the other KD(P)G aldolases of known structure, the Pt-KDG-aldolase has a distinct preference for KDG over KDPG as a substrate, the catalytic efficiency with KDG being 2,000 times that with KDPG.4 To investigate this specificity in the outer part of the binding pocket, the structure of Pt-KDG-aldolase was compared to that of Ss-KD(P)G-aldolase in complex with KDPG (Fig. 5B). None of the Ss-KD(P)G-aldolase phosphate-interacting residues is conserved in Pt-KDG-aldolase, the site of Arg237 being occupied by glycine (241), and Ser241 by phenylalanine (245), neither of which will form constructive interactions with a phosphate moiety. The main chain of Arg106 of Ss-KD(P)Galdolase aligns with Lys107 in Pt-KDG-aldolase, also a positive amino acid, but the latter forms an interaction within its own subunit, and is not available for substrate interaction (Fig. 5A). The entrance to the active site surface is considerably more hydrophobic in Pt-KDG-aldolase (Fig 5C), although some positive charge is provided by Arg166 from the subunit on the other side (Fig. 5A), which could make longer-distance interactions with one phosphate conformation. However the other phosphate conformation would be unfavourable due to clashes, particularly with Phe 245, and the hydrophobic nature of these outer reaches of the Pt-KDG-aldolase catalytic pocket is increased further by Ile133 (Pt) instead of Tyr132 (Ss), and Phe136 (Pt) replacing Ala135 (Ss) (Fig. 5C,D). These differences are predicted to explain the lower affinity of Pt-KDG-aldolase for phosphorylated substrates.

Site-directed mutagenesis To investigate these active-site observations further, a series of residues was swapped by sitedirected mutagenesis and the resulting kinetic analyses of the variant enzymes are summarised in Table 2; kinetic analyses from selected variants are shown graphically in Supplementary Figures S1 and S2. Initial studies concentrated on three of the residues in Ss-KD(P)G-aldolase that interact with the phosphate group of the substrate KDPG, namely Arg237, Arg106 and Tyr132 . Thus Ss-KD(P)Galdolase R237G raised the KM for KDPG approximately 5-fold with no significant change in the Vmax. There were no significant changes in the kinetic parameters with KDG as substrate. The reverse change in Pt-KDG-aldolase, G241R, increased the kcat/KM for KDPG 9-fold but, surprisingly, this came


Biochemistry 1 2 360 3 4 361 5 6 362 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 41

from an increase in kcat rather than an expected decrease in the KM; as predicted, no change was observed in the kinetic parameters with KDG as substrate.


Page 21 of 41 1 2 363 3 4 364 5 6 365 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 36 366 37 38 39 367 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Table 2: Kinetic parameters determined for mutated Ss and Pt aldolases.

S. solfataricus KD(P)G Aldolase KDPG as substrate

Vmax (U/mg)

KM (mM)

kcat (s-1)

kcat/KM (s-1M-1)

Native enzyme

168.4 ± 17.3

0.69 ± 0.09

97.8 ± 10.1

1.42 (± 0.24) x 105

R237G

174.4 ± 9.3

3.28 ± 0.28

101.1 ± 5.4

0.31 (± 0.03) x 105

R106I

157.6 ± 9.2

0.98 ± 0.15

91.4 ± 5.3

0.93 (± 0.15) x 105

Y132I

332.1 ± 43.8

0.68 ± 0.03

192.6 ± 25.3

2.83 (± 0.39) x 105

R106I/Y132I

141.6 ± 13.4

1.57 ± 0.33

82.1 ± 7.8

0.52 (± 0.12) x 105

S241F

200.4 ± 6.6

0.89 ± 0.08

116.2 ± 3.8

1.31 (± 0.13) x 105

S241F/L242P

114 ± 4.2

0.58 ± 0.07

66.1 ± 2.4

1.14 (±0.14) x 105

R106I/Y132I/S241F

85.3 ± 3.9

2.45 ± 0.20

49.5 ± 2.3

0.20 (± 0.02) x 105


Biochemistry 1 2 368 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 369 27 28 370 29 30 371 31 32 372 33 34 373 35 36 374 37 38 39 375 40 41 376 42 43 377 44 45 378 46 47 379 48 49 380 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 41

P. torridus KDG Aldolase Vmax (U/mg)

kcat (s-1)

KM (mM)

kcat/KM (s-1M-1)

KDPG as substrate Native enzyme

0.785 ± 0.07

6.4 ± 1.5

0.398 ± 0.035

62.14 ± 15.6

G241R

6.77 ± 0.56

6.27 ± 0.9

3.43 ± 0.283

546.98± 90.6

KDG as substrate Native enzyme

62.4 ± 11.6

1.12 ± 0.3

31.6 ± 5.88

2.82 (± 0.91) x 104

G241R

63.4 ± 9.4

1.06 ± 0.08

32.12 ± 4.78

3.03 (± 0.51) x 104

In each active site of Ss-KD(P)G-aldolase, Arg106 from a neighbouring subunit interacts with the phosphate group of KDPG, and accordingly the mutation R106I raised the KM for KDPG, although the effect was much smaller (1.4-fold) than that for R237G. The residue in the equivalent position in the Picrophilus enzyme is Lys107; attempts were therefore made to reposition this Lys to one similar to Arg106 in Ss-KD(P)G-aldolase. Using the sequence alignment as a guide (Fig. 6), a Pro was inserted in Pt-KDG-aldolase before Lys107, and Phe245 was changed to Ser245 as the bulky Phe appeared to be blocking the binding of the phosphate of KDPG; unfortunately, the enzyme lost over 70% of the wild-type enzyme activity, and size exclusion chromatography indicated that the enzyme was dissociating. The F245S mutant without the insertion of the Pro also had no defined oligomeric structure.


Page 23 of 41 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 29 381 30 382 31 32 383 33 34 384 35 36 385 37 38 386 39 40 387 41 42 388 43 44 45 389 46 47 390 48 49 391 50 392 51 52 393 53 54 55 394 56 57 395 58 59 60

Biochemistry

Fig 6: Structure-based sequence alignment (calculated with Expresso24) and displayed with Espript25 of Pt-KDG-aldolase (KDGA_PICTO), Thermoplasma acidophilum KDG-aldolase (DAPAL_THEAC), S. acidocaldarius KD(P)G-aldolase (KDGA_SULAC), Ss-KD(P)G-aldolase (KDGA_SULSF) and T. tenax KD(P)G-aldolase (KDGA_THETE), with the sequences labeled with their Uniprot codes and the secondary structure depicted above the sequences. Identical amino acids are in white text on a red background; similar amino acids are in red text. The four amino acids labeled in cyan are those shown to interact with the phosphate group of KDPG in Ss-KD(P)G-aldolase, R106, Y132, R237 and S241 (Ss numbering). Those in orange are thought to make phosphate binding less likely I133, G241, F245 (Pt numbering).

Surprisingly, mutation of Y132 to Ile did not affect the KM of Ss-KD(P)G-aldolase for KDPG, but increased kcat 1.9 fold; the reason for this is perhaps that in the absence of Y132, which interacts with conformation A, the phosphate preferentially occupies the other position in the active site and the


Biochemistry 1 2 396 3 4 397 5 6 398 7 8 399 9 10 400 11 12 401 13 14 15 402 16 17 403 18 19 404 20 21 405 22 23 406 24 25 407 26 27 408 28 29 409 30 31 410 32 33 34 411 35 36 412 37 38 413 39 40 414 41 42 415 43 44 416 45 46 417 47 48 418 49 50 419 51 52 53 420 54 55 421 56 57 422 58 59 60

Page 24 of 41

replacement with the smaller Ile allows a more open active site. However, the mutation of Y132 in combination with R106I, affecting both phosphate binding sites, did increase the KM two-fold over R106I alone. Mutation of I133Y in Pt-KDG-aldolase, to mimic the Ss-KD(P)G-aldolase Y132, was attempted but caused dissociation of the oligomeric structure with considerable loss of catalytic activity. Ser241 forms a hydrogen bond with the phosphate of KDPG in Ss-KD(P)G-aldolase, whereas this position is occupied by a Phe in Pt-KDG-aldolase, which will not form hydrogen bonds and appears to obstruct the access of the phosphorylated substrate to the active site, particularly in one of the two KDPG conformations. However, the mutation S241F in Ss-KD(P)G-aldolase had only a small effect on the catalysis of KDPG. This may be because only one of the two phosphate binding conformations is affected; moreover catalysis with KDG as substrate was similarly unaffected. Also, in Pt-KDG-aldolase the Phe-containing loop takes up a different conformation from that containing S241 in Ss-KD(P)G-aldolase, possibly due to the Pro that lies underneath the Phe; consequently, the double mutant, S241F/L242P, aimed to allow the Phe to take up the Picrophilus-like conformation, but again little effect on the binding of KDPG was observed. However, the influence of Ser241 on the binding of KDPG was best demonstrated in the triple mutant R106I/Y132I/S241F, where a 4-fold increase in KM was observed; that is, doubling the increase seen in the double mutant R106I/Y132I. Thus, Ser241 contributes to KDPG binding as part of a network with R106 and Y132 and together with R237, constitutes the fourth amino acid of the network.

Concluding remarks In this communication we report the 3D structure of Pt-KDG-aldolase, which is highly specific for KDG, and the structure of Ss-KD(P)G-aldolase in complex with KDPG, where the phosphate was unexpectedly shown to bind in two conformations. By structural comparison of Pt-KDG-aldolase and KDPG-cleaving Ss-KD(P)G-aldolase, and by site-directed mutagenesis studies, a network of four amino acids composed of Arg 106, Tyr 132, Arg 237 and Ser241 was identified that contacts the phosphate of KDPG in Ss-KD(P)G-aldolase (Figure 7). This KDPG-binding network was absent in Pt-


Page 25 of 41 1 2 423 3 4 424 5 6 425 7 8 426 9 10 427 11 12 428 13 14 429 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 44 430 45 431 46 47 432 48 49 433 50 51 434 52 53 435 54 55 436 56 437 57 58 59 60

Biochemistry

KDG-aldolase, explaining the low catalytic efficiency of this aldolase for KDPG. The respective substrate specificities of the two aldolases for KDG and KDPG determine the two different routes of glucose degradation via non-phosphorylative or branched Entner-Doudoroff pathway in these two thermophilic archaea (Fig 1). The operation of the non-phosphorylative Entner-Doudoroff pathway in P. torridus is further supported by the absence of a KDG-kinase gene in this organism.

Fig 7. Interactions (magenta dotted lines with lengths in Å) of a four amino acid-network (Arg106, Tyr132, Arg237 and Ser241) predicted from structural data and shown through site-directed mutagenesis to be important for binding the phosphate group of KDPG in Ss-KD(P)G-aldolase. SsKD(P)G-aldolase amino acids from one monomer in cyan, from a neighbour in dark blue, and KDPG in yellow). The presence or the absence of the four amino acid network in sequences of homologous


Biochemistry 1 2 438 3 4 439 5 6 440 7 8 441 9 10 442 11 12 443 13 14 15 444 16 17 445 18 19 446 20 21 447 22 448 23 24 449 25 26 450 27 28 451 29 30 452 31 32 453 33 34 35 454 36 37 455 38 39 456 40 41 457 42 43 458 44 45 459 46 47 460 48 49 461 50 51 462 52 53 54 463 55 56 464 57 58 59 60

Page 26 of 41

aldolases enables the prediction of substrate specificities of these aldolases and their functional involvement in archaeal Entner-Doudoroff pathways. The presence of the network in aldolases from Sulfolobus acidocaldarius and Thermoproteus tenax (see structure based alignment in Fig 6.) is in accordance with the kinetic properties showing high preference for KDPG over KDG; both organisms have been reported to degrade glucose via a branched Entner-Doudoroff pathway .4,6 Furthermore, the absence of the network in the aldolase from Thermoplasma acidophilum is in accordance with the high specificity for KDG over KDPG and with the operation of the non-phosphorylative EntnerDoudoroff pathway in this organism.4

Author contributions: MR and MO produced the proteins for crystallisation. VZ crystallised Pt-KDGaldolase and with GLT collected X-ray data on Pt-KDG-aldolase crystals; SJC solved and refined the Pt-KDG-aldolase crystal structure and also crystallised and solved the structure of the Ss-KD(P)Galdolase complex with KDPG.

MO and UJ performed the mutagenesis experiments and MJD, PS,

and UJ analysed the enzyme kinetic data . PS initiated the project and SJC, PS, MJD and UJ have been involved in writing the paper.

Acknowledgements Initial work of site-directed mutagenesis studies was performed by Jasmin Gille as part of her bachelor thesis (Christian-Albrechts-Universität Kiel). Refinement of the Pt-KDG-aldolase structure was carried out as part of the work submitted by Alexandros Priftis for an MRes in Protein Structure and Function at the University of Bath. The authors would like to thank Diamond Light Source for beamtime (proposal mx17212-8), and Gyles Cozier, William Bradshaw, Shalini Iyer and the staff of beamline i04 for assistance with data collection. This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. None of the authors have any real or perceived conflicts of interest.


Page 27 of 41 1 2 465 3 4 466 5 6 467 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. Kinetic analyses of wild-type and mutant aldolases from Sulfolobus (Figure S1) and Picrophilus (Figure S2).


Biochemistry 1 2 468 3 4 469 5 6 470 7 8 471 9 10 11 472 12 13 473 14 15 16 474 17 18 475 19 20 476 21 22 477 23 24 25 478 26 27 479 28 29 480 30 31 481 32 33 34 482 35 36 483 37 38 484 39 40 41 485 42 43 486 44 45 487 46 47 48 488 49 50 489 51 52 490 53 54 491 55 56 57 492 58 59 60

Page 28 of 41

References 1. Bräsen, C.; Esser, D.; Rauch, B.; Siebers, B. (2014) Carbohydrate metabolism in Archaea: current insights into unusual enzymes and pathways and their regulation. Microbiol. Mol. Biol. Rev. 78, 89175. 2. Danson, M.J.; Lamble, H.J.; Hough, D.W. (2007) Central metabolism. in Archaea: Molecular and Cellular Biology (Cavicchioli R. Ed.), p 260-287, ASM Press, Washington DC. 3. Siebers, B.; Schönheit, P. (2005). Unusual pathways and enzymes of central carbohydrate metabolism in Archaea. Curr. Opin. Microbiol. 8, 95-705. 4. Reher, M.; Fuhrer, T.; Bott, M.; Schönheit, P. (2010) The nonphosphorylative Entner-Doudoroff pathway in the thermoacidophilic euryarchaeon Picophilus torridus involves a novel 2-keto-3deoxygluconate-specific aldolase. J. Bacteriol. 192, 964-974. 5. Angelov, A.; Fütterer, O.; Valerius, O.; Braus, G.H.; Liebl, W. (2005) Properties of the recombinant glucose/galactose dehydrogenase from the extreme thermoacidophile, Picrophilus torridus. FEBS J. 272, 1054-1062. 6. Ahmed, H.; Ettema, T.J.G.; Tjaden, B.; Geerling, A.C.M.; van der Oost, J.; Siebers, B. (2005) The semi-phosphorylative Entner-Doudoroff pathway in hyperthermophilic archaea - a re-evaluation. Biochem J. 390, 529-540. 7. Lamble, H.J.; Theodossis, A.;. Milburn, C.C.; Taylor, G.L.;. Bull, S.D.; Hough, D.W.; Danson, M.J. (2005) Promiscuity in the part-phosphorylative Entner-Doudoroff pathway of the archaeon Sulfolobus solfataricus. FEBS Lett. 579, 6865-6869. 8. Lamble, H.J.; Heyer, N.I.; Bull, S.D.; Hough, D.W.; Danson, M.J. (2003) Metabolic pathway promiscuity in the archaeon Sulfolobus solfataricus revealed by studies on glucose dehydrogenase and 2-keto-3-deoxygluconate aldolase. J. Biol. Chem. 278, 34066-34072. 9. Wolterink-van Loo, S.; van Eerde, A.; Siemerink, M.A.J.; Akerboom, J.; Dijkstra, B.W.; van der Oost, J. (2007) Biochemical and structural exploration of the catalytic capacity of Sulfolobus KDG aldolases.


Page 29 of 41 1 2 493 3 4 5 494 6 7 495 8 9 496 10 11 497 12 13 498 14 15 16 499 17 18 500 19 20 501 21 22 502 23 24 25 503 26 27 504 28 29 505 30 31 506 32 33 507 34 35 36 508 37 38 509 39 40 510 41 42 43 511 44 45 512 46 47 48 513 49 50 514 51 52 515 53 54 516 55 56 57 517 58 59 60

Biochemistry

Biochem. J. 403, 421-430. 10. Sutter, J.M.; Tästensen, J.B.; Johnsen, U.; Soppa, J.; Schönheit, P. (2016) Key enzymes of the semiphosphorylative

Entner-Doudoroff

pathway

in

the

haloarchaeon

Haloferax

volcanii:

characterization of Glucose Dehydrogenase, Gluconate Dehydratase, and 2-Keto-3-Deoxy-6Phosphogluconate Aldolase. J. Bacteriol. 198, 2251-2262. 11. Allard, J.; Grochulski, P.; Sygusch, J. (2001) Covalent intermediate trapped in 2-keto-3-deoxy-6phosphogluconate (KDPG) aldolase structure at 1.95-Å resolution. Proc. Natl. Acad. Sci. USA 98, 3679-3684. 12. Pauluhn, A.; Ahmed, H.; Lorentzen, E.; Buchinger, S.; Schomburg, D.; Siebers, B.; Pohl, E. (2008) Crystal structure and stereochemical studies of KD(P)D aldolase from Thermoproteus tenax. Prot. Struct. Func. Bioinf. 72, 35-43. 13. Theodossis, A.; Walden, H.; Westwick, E.J.; Connaris, H.; Lamble, H.J.; Hough, D.W.; Danson, M.J.; Taylor, G.L. (2004) The structural basis for substrate promiscuity in 2-keto-3-deoxygluconate aldolase from the Entner-Doudoroff pathway in Sulfolobus solfataricus. J. Biol. Chem. 279, 4388643892. 14. Otwinowski, Z.; Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation modein Methods Enzymol. 276 Macromolecular Crystallography, Part A (Carter, C.W. Jr, Sweet, R.M. Ed), , p307-326, Academic Press, New York. 15. Long, F.; Vagin, A.A.; Young, P.; Murshudov, G.N. (2008) BALBES: a molecular-replacement pipeline. Acta Crystallogr. D64, 125-132. 16. Blagova, E.; Levdikov, V.; Milioti, N.; Fogg, M.J.; Kalliomaa, A.K.; Brannigan, J.A.; Wilson, K.S.; Wilkinson, A.J. (2006) Crystal structure of dihydrodipicolinate synthase (BA3935) from Bacillus anthracis at 1.94Å resolution. Prot. Struct. Func. Bioinf. 62, 297-301. 17. Emsley, P.; Lohkamp, B.; Scott, W.; Cowtan, K. (2010). Features and Development of Coot. Acta Crystallogr. D 66, 486-501.


Biochemistry 1 2 518 3 4 519 5 6 520 7 8 521 9 10 11 522 12 13 523 14 15 524 16 17 18 525 19 20 526 21 22 527 23 24 25 528 26 27 529 28 29 30 530 31 32 531 33 34 532 35 36 37 533 38 39 534 40 41 535 42 43 536 44 45 46 537 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 41

18. Adams, P.D.; Afonine, P.V.; Bunkoczi, G.; Chen, V.B.; Davis, I.W.; Echols, N.; Headd, J.J.; Hung, L.W.; Kapral, G.J.; Grosse-Kunstleve, R.W.; McCoy, A.J.; Moriarty, N.W.; Oeffner, R.; Read, R.J.; Richardson, D.C.; Richardson, J.S.; Terwilliger, T.C.; Zwart, P.H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D 66, 213-221. 19. Winter, G.; Waterman, D.G.; Parkhurst, J.M.; Brewster, A.S.; Gildea, R.J.; Gerstel, M.; FuentesMontero, L.; Vollmar, M.; Michels-Clark, T.; Young, I.D.; Sauter, N.K.; Evans, G. (2018) DIALS: implementations and evaluation of a new integration package. Acta Crystallogr. D74, 85-97. 20. Evans, P.R.; Murshudov, G.N. (2013) How good are my data and what is the resolution?. Acta Crystallogr. D69, 1204-14. 21. Lebedev, A.A.; Young, P.; Isupov, M.N.; Moroz, O.V.; Vagin, A.A.; Murshudov, G.N. (2012) Jigand: A graphical tool for the CCP4 template-restraint library. Acta Crystallogr. D68, 431-440. 22. The PyMOL Molecular Graphics System, Version 1.8 Schrödinger, LLC. .2015. 23. Krissinel, E.; Henrick, K. (2007) Inference of macromolecular assemblies from crystalline state. J Mol. Biol. 372, 774-797. 24. Armougom, F.; Moretti, S.; Poirot, O.; Audic, S.; Dumas, P.; Schaeli, B.; Keduas, K.; Notredame, C. (2006) Expresso: automatic incorporation of structural information in multiple sequence alignments using 3D-Coffee. Nucleic Acids Res. 34, W604-608. 25. Robert, X.; Gouet, R. (2014). Deciphering key features in protein structures with the new ENDscript server. Nucleic Acids Res. 42: W320-W324.


Page 31 of 41 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: The structure of Pt-KDG-aldolase in cartoon representation, with each monomer in a different colour. The Ramachandran-outliers Tyr 105 are shown in stick form, and can be seen to protrude into a neighbouring molecule as part of the larger dimer interface, the perpendicular tetramer interface being less extensive. 54x40mm (300 x 300 DPI)

ACS Paragon Plus Environment

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

Fig 3. Cartoon representation of Ss-KD(P)G-aldolase (grey) aligned to Pt-KDG-aldolase (green) to show the overall structural similarity. The active site is marked by a stick representation of the Lysine (155 and 159 respectively) which forms the Schiff base in the catalytic reaction, and which is bound to KDPG (yellow) in the Ss-KD(P)G-aldolase complex structure. The three N-terminal amino acids of the Pt-KDG-aldolase are removed for clarity. 47x40mm (300 x 300 DPI)


Page 32 of 41

Page 33 of 41 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 4A 81x60mm (300 x 300 DPI)


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

Fig 4. Following superimposition of the structures, the active sites of (A) the apo structure of Pt-KDGaldolase (GOL are glycerol molecules) and (B) Ss-KD(P)G-aldolase with KDPG (yellow bonds) bound are shown in the same orientation with the 2FoFc electron density map for each structure, contoured at 1.5σ restricted to that around KDPG in B for clarity. B also contains the structure of Ss-KD(P)G-aldolase with KDG bound (grey) superimposed to show the relative stasis of all but those amino acids that interact with the phosphates. The main chain cartoon containing Arg106 from a neighbouring monomer is omitted as it crosses the middle of the image. Hydrogen bonds are shown as dotted lines.

81x60mm (300 x 300 DPI)


Page 34 of 41

Page 35 of 41 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 5A 81x60mm (300 x 300 DPI)


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

Figure 5B 81x60mm (300 x 300 DPI)


Page 36 of 41

Page 37 of 41 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 5C 81x60mm (300 x 300 DPI)


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

Fig 5: Analysis of interactions (shown in magenta dotted lines) made by KDPG with the active sites of PtKDG-aldolase (modeled on the complex between KDPG and Ss-KD(P)G-aldolase) (A,C), and those seen in Ss-KD(P)G-aldolase (B,D). (A) Stick representations of amino acids from Pt-KDG-aldolase (green) that could interact with KDPG (magenta), with those amino acids from neighbouring subunits in different shades of green. (B) A similar representation of amino acids in the Ss-KD(P)G-aldolase active site, with those from a neighbouring monomer in a different shade of blue. (C) Electrostatic surface around the catalytic site of Pt-KDG-aldolase, with positively-charged areas in blue and negative in red. (D) Electrostatic surface around the catalytic site of Ss-KD(P)G-aldolase, coloured similarly. 81x60mm (300 x 300 DPI)


Page 38 of 41

Page 39 of 41 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

KDGA_PICTO DAPAL_THEAC KDGA_SULAC KDGA_SULSF KDGA_THETE




Biochemistry

1 1 1 1 1

66 66 64 65 83

144 146 143 144 162

226 229 220 223 241

jjjjk iiiiiiiiiiiii jjk lll iiiiiiiiiiiii jjjjk iiiiiiiiiiiii jjjk lll iiiiiiiiiii jjjjk iiiiiiiiiiiii jjjk lll lll iiiiiiiiiiii jjjjk jk jk iiiiiiiiiiiii jjjk lll X X X X X X iiiiiiiiiiiii X MYK....................GI c DF VCP c DEEEF MI cc TPLDAHG A NI DF DYNA A TNILI A KY DF LEGIN DEF VD c YL DF F c PMG c DEEEF ST cc GVFP c A YFT A LKE c DF RKDFL ABKG FF VVABRR EDAB MYK....................GI c KG VTP c KdddG MI cc TPMGQNG B EI KG DYRA B TEILI B DN KG LADFG KdG VD c GL KG F c PMG c KdddG ST cc GLFP c B MFS B TDE c KG KKKFL ..M....................EI cKG KG ISP ccII IIcc KdddG TPFDKQG BK VKGND VK DE AKBBLL KK TI HH AABBKE NNKG LLEKG KdG ID ccAK ILKG F cV N GKdddG ccTT TTcc GLGP BcA L SBK D EKG ccKK RL QE NN LLBBNK AA LVBBYY DDBB .MP....................EI cJH ID ..MSHPLIPFRANFLRAPRVLSMEI cIV TA PPKdddG cV Icc JIIIH ccTT PT FF TR K. DG NGBCRR ILKG JHD P E LCF A N H VCK NKG JHLI IT RS KK GGKdG JIH VD cV VKG JHFFccVV NA GGKdddG cT Tcc JIIIH ccGG LL GG PPBcCcSA LL SSBCPL EQ EEKG cK M E L TCD A ACT SC JH iiiiiiiiiiiii jjjk iiiiiiiiiiii jjjjjk ii jj. . jjk jjk iiiiiiiiiiii jjjk iiiiiiiiiiiii jjjjk iiii ii ..i iiiiiiiiiilll jjjjk iiiiiiiiiiiii jjjjk iiii ii ..i i Xj.. .. jjk jjk iiiiiiiiiiiii jjjk iiiiiiiiiiiii i jjjjk iiii X X X X X .. X X ii NS DF KKPI AB.E .V MYABAA GG VVDEF cKdG GSSS DF INE DEF VNEL cKdddG DEEEF MK ccFY ST ME DDABII GG IIABEK AV ARDF VLMPP DEF Y cc YI.K DF LNQ ABED AW IMABYY HRDEF YYKEI ABLI SS .A .A SS NNABDD M. D. LLDEEEF LIY cccc NIP ABQQ FLDF T.NK.I ABDS PE ES HS KG KK.I cKdG GSSS KG TQE KdG SVEL cKdddG SK KG VLMPT KdG Y cc YI.K KG PDE KdG HFSTV KdddG FIY cccc NIP KG S.GSWI VT KG H.KL B ..I B FQV c GSLN KG LND KdG VMEL c VK c FSNE B MDI B LGV KG SSHSP KdG Y cc YFPR c KG LPE B KFL B AK KdG YYEEI B AR..IS B SHSL KdddG YIY cccc NYP B AA KG TGYD.I BBPD PA SK VT KG N.KI B ..I B FQV c KdG GGLN KG LDD KdG AIRL c KdddG AK c LSKD B FDI B VGI KG ASYAP KdG Y cc YYPR c KG MSE B KHL B VK KdG YFKTL B CE..VS B PHPV KdddG YLY cccc NYP B TA KG TGKD.I AA JHR . R VC. . ICV Q VJIH cA S L NJHA D EJIH AIAL cA KcY A E SCR G ACE A VJHA S L P PJIH JIIIH Y ccY F P RJH cL S ECR Q ICA KJIH YFRDL CC S . . A VCS I P VJIIIH FLY cccc NYP CA AJHV G R D . VCD A R iiiiiiii jjjjjk iiiiiiii jjjk llliiiiii jjk llll iiiiiiii iiii ii jjjjjk iiiiiiiiii jjjk llliiiii jjk llll iiiiiiiiiiii iiii iiiii jjjjjk iiiiiiiiii jjjk llliiiiii jjk llll iiiiiiiiiiii iiii iiiii X jjjjjkX iiiiiiiiii jjjk llliiiiii jkX lll iiiiiiiiiiii X X X X X iiii DEF T VKNL ABKT SR EE FF SS SN VVDEEEEEEF KG ccVI KKcc DSSADI ABRR GF FF MSABER MI LI SR LH SK DN DE DK FF AD VIDEEEEF FQG ccQQ DD DD LLDEF cKdG LFTS DF LELGA ABSS GGAcBcGG VVAcBcCC GG TL TSDEF N ccFI SS .. DPABGY IIDF VRL ABYY HH EE YF KS .A NG NN .. RLDF EMA AcBc KdG L IEKL KddddddG KG cc DSSGDM KddddG FQG cKdG LFLS KG LSIGA KdG N KG TNL KG EKA KdG I LKSL B .....PV KddddddG KG c IK cc DTNQDL B AHSL B EYKLNLPG.VKV KddddG YNG c SNTL c IYYS KG LLS.L B DG c B VV c B ASFT KdG N c FIPE B VI KG VKQ B RDLIK.QGKL KG DDA BcBc KdG V AKEI B ....GCF KddddddG TG c VK cc DTIENI B IHTL B DYKRLNPN.MLV KddddG YSG c SDML c KdG IATV KG ASTGL B DG c B NV c B AAGS KdG N c YLPE B VT KG VTI B KKLAM.ERKI KG DEA JIH A AKEL C. . . . G C IJIIIIIIH RG cV KccD T N E S LCA H T LCA Y K R Y L P Q . A R VJIIIIH YNG cS D S LJIH cV F A SJHF A V R LCD GCcV VCcA S S AJIH N cY L P ECL LJHA G ICR D A V A . A G D IJHE R ACc iiiiii iiiiiiiii iiiiiiii iiiiii. iiii iiiii iiiiiiiiiiiiiiiiiiiii . iiiiiiiiii . iiiii iiiii iiiiiiiiii iiiiiiii iiiiiiiiiiiiiiiiiiii . . iiiii iiiiiX. iiiiiiiiiiii X iiiiiiii X . iiiii X X iiiiiiiiii iiiii LK DEF IEKNDV ABIN PP LLDF MKKL ABGNccKE YA QT FFccPP NA AG YY YYAcBcE. YY .A FF YY KK KM NNDF NING.GYR cKdddG DEEEF PP ccMM YV .E .P .T .T .D .Q .K .K RK VIABGD IQ EEAcBcIL .T .K .I .P .K .K .Q .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. RK KdG IQIDEV KG MYAI KG GIKG.GYR c AP LR KdG LQE.LI BBND RE LVKG ADIL BRcK Y G ScI S A I YBcV L . V N E F QKGGG YY DD VL GG YY PP RRKdddG ccPP PPccII FF .. .. .. .. .. .. .. .. PP LLBBTD DD EEBcBcEE AE LR SQ LL KI RK EK IV EE PG LI KR RA KK IL QV EE LL VK HI .L .K .E LK KdG LQF.LH RS JIH LQF.LL CD E IKG JHIV EE AS SABCRRccIH FI GG SYccLA SA SA NV YYBcCcVE LL .. TV KE YI FF QQKG JHG Y E A G E P RKdddG cG PcV Y . . . . . . . . P LCD P ECcE K A W L R A A V A K A K S Q L R L . . . . . . JIIIH


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

Fig 7. Interactions (magenta dotted lines with lengths in Å) of a four amino acid-network (Arg106, Tyr132, Arg237 and Ser241) predicted from structural data and shown through site-directed mutagenesis to be important for binding the phosphate group of KDPG in Ss-KD(P)G-aldolase. Ss-KD(P)G-aldolase amino acids from one monomer in cyan, from a neighbour in dark blue, and KDPG in yellow). 44x44mm (300 x 300 DPI)


Page 40 of 41

Page 41 of 41 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 (Figure 7 rotated) 60x33mm (300 x 300 DPI)


Insights into substrate specificity of archaeal Entner ... - ACS Publications

Recommend Documents