Figures and data
UDP-GlcA coordination and proofreading.
A, CryoEM density map of CvHAS bound to UDP-GlcA with a grey bar representing the membrane boundary. UDP-GlcA density shown in light sea green and yellow for its UDP and GlcA moieties, respectively. B-C, Coordination of an inserted UDP-GlcA substrate in the absence of a GlcNAc primer. UDP-GlcA is shown as a ball and stick model with light sea green carbon atoms for the UDP moiety and yellow carbon atoms for GlcA. D, CryoEM density for Arg341 rotamer 1 (R1) and rotamer 2 (R2). E, Coordination of UDP-GlcA in the proofreading conformation. F, CryoEM density for UDP-GlcA’s sugar ring and surrounding residues in the proofreading state. G-H, Alignment of proofreading and inserted UDP-GlcA positions. The ligand in the inserted UDP-GlcA structure is shown in grey. H, Activity of uracil binding pocket mutants for CvHAS. Activity measurements were normalized to wild-type (WT), and were reported as the average of three technical replicates. Error bars represent the standard deviation from the mean.
Biochemical analysis of UDP-GlcA and UDP-GlcNAc utilization by CvHAS.
A-B, Binding isotherms derived from ITC experiments where CvHAS was titrated with UDP-GlcA in the absence (A) or presence (B) of GlcNAc. C, Scatter plot of UDP-GlcNAc turnover without an acceptor (▪) or in the presence of 2.5 mM UDP-GlcA (▴). D, UDP-GlcA turnover without an acceptor (◆), in the presence of 10 mM GlcNAc (▪) and in the presence of 2.5 mM UDP-GlcNAc (▴). Data points for panels C and D are reported as the average of five technical replicates. E, Measurement of UDP-GlcA turnover at a constant concentration of 2.0 mM with titration of GlcNAc (▪). Data points were reported as the average of three technical replicates. F, HA electrophoresis gel analyzing HA production under UDP-GlcNAc limiting conditions with excess UDP-GlcA . All error bars represent the standard deviation from the mean. G, HA electrophoresis gel analyzing HA production under UDP-GlcA limiting conditions with excess UDP-GlcNAc. Confirmation of HA product identity by HA lyase digestion under the provided synthesis conditions has been described in previous reports10.
DDM binding at the acceptor site.
A, Left - Cross-section of a cryoEM density map for CvHAS (blue) bound to DDM (green). Middle and right – atomic coordinates of CvHAS with DDM cryoEM density independently contoured. Relevant interfacial (IF) and transmembrane (TM) helices are labeled. B, DDM coordination. DDM is shown in ball and stick representation with light green carbon atom coloring. C, Structural alignment of DDM-bound and HA disaccharide (HA2)-bound (PDB ID: 8snc) coordinates. The switch loop (266-271) is displayed with backbone atoms only for clarity, with the DDM-bound conformation shown in light blue and the HA2-bound conformation shown in grey. A solid arrow is used to indicate direction of switch loop flipping from the HA2 bound to DDM bound state. A dashed arrow is used to indicate the direction of displacement of DDM’s maltose group relative to the HA disaccharide.
Transfer of GlcA to non-canonical acceptor substrates.
A – B, Relative reaction rates for UDP-GlcA (A) and UDP-GlcNAc (B) turnover in the presence of GlcNAc, chitobiose (Chi2) and cellobiose (Cel2). Rates were normalized to samples where the acceptor volume was replaced with water (light grey). Individual velocities were calculated as the average of three technical replicates. Error bars correspond to the standard deviation from the mean. C,D – Diphenylamine staining of TLC plates developed after spotting CvHAS glycosyl transfer reaction mixtures containing either UDP-GlcA (C) or UDP-GlcNAc (D).
CryoEM data processing for UDP-GlcA inserted and proofreading structures.
Carves of cryo electron density maps for UDP-GlcA in inserted and proofreading states are displayed as a black mesh. Local resolution maps are colored according to estimated resolution at FSC = 0.143 in Å.
CryoEM density for Mn and the priming loop c-terminus in inserted and proofreading conformations.
A, Cross-section of the cryoEM density map for the inserted UDP-GlcA pose. CvHAS and ligand carbon atoms are colored in light cyan, with the ligand represented as a ball and stick model. B, Cross-section of cryoEM density map for the proofreading UDP-GlcA pose. Protein and ligand carbon atoms are colored in grey. C-D Active site view showing local cryoEM density for putative manganese ions (purple) in inserted (C) and proofreading (D) UDP-GlcA bound CvHAS. E, Active site view showing continuous cryoEM density for Gly300 in the inserted UDP-GlcA bound state. F, Broken cryoEM density due to an unresolved Gly300 in the proofreading UDP-GlcA bound state.
Purification CvHAS’ uracil pocket mutants.
A-E, S200 Increase chromatography of WT CvHAS (A), Y91A (B), Y91F (C), H174A (D) and H174W (E). Peaks pooled for subsequent biochemistry are indicated with an asterisk (*). F, Coomassie stained SDS-PAGE gel for purified XlHAS-1, WT CvHAS, and CvHAS mutants.
ITC plots for CvHAS substrate titration.
A-B, raw ITC plots for UDP-GlcA titration in the absence (A) or presence (B) of excess GlcNAc. C-E, Raw ITC plots for UDP-GlcNAc, UDP-Glc and UDP.
Michaelis-Menten fits for CvHAS substrate titration.
A, CvHAS titration with UDP-GlcNAc in the absence of an acceptor. B, Titration of UDP-GlcA in the presence of 10 mM GlcNAc. C, Titration of GlcNAc in the presence of 2.0 mM UDP-GlcA. Five technical replicates were used to derive average reaction rates and standard deviations for plotting of UDP-GlcNAc and UDP-GlcA titrations. Three technical replicates were used for fitting the GlcNAc titration series. All non-linear regressions were performed in Prism 6.0. Error bars represent standard deviations from the means.
HA synthesis activity is abolished by DDM.
Measurements of HAS activity were taken as the average of three technical replicates. Activity values for CvHAS and XlHAS were independently normalized to the control condition. Error bars correspond to the standard deviation from the mean.
CryoEM Data processing for DDM bound CvHAS.
CryoSPARC data processing workflow for CvHAS bound to DDM. Carves of DDM density are shown as a black mesh. Local resolution estimates calculated at FSC = 0.143 are reported in Å.
Switch loop movement in GlcNAc, HA disaccharide and DDM-bound CvHAS structures.
A, Superimposed structures for the GlcNAc primed, UDP-GlcA-bound CvHAS (grey) and DDM-bound CvHAS (blue). Switch loop movement is indicated by a black arrow. B-C, Position of the switch loop and interactions with GlcNAc in primed, UDP-GlcA-bound (PDB ID: 8snd) CvHAS (B) and HA disaccharide-bound (PDB ID: 8snc) CvHAS (C).
Kinetic traces of UDP-GlcA hydrolysis.
Individual kinetic traces for UDP-GlcA turnover in the presence of potential non-canonical glycosyl transfer acceptors supplemented at 12.5, 25, and 50 mM concentrations. The window used for determining reaction velocities is indicated by two vertical dashed lines.
Extension of cellobiose and chitobiose by GlcA.
A, TLC analysis of cellobiose (Cel2) titration in the presence of excess UDP-GlcA and CvHAS. Staining was performed with thymol reagent. B, TLC analysis of chitobiose (Chi2) titration. In the presence of excess UDP-GlcA and CvHAS. Staining was performed with diphenylamine reagent. C, Autoradiograph of TLC experiment measuring transfer of 14C-GlcA to cellobiose and chitobiose.
Comparison of UDP-GlcA binding by CvHAS to UDP-GlcNAc binding by CHS.
A, Inserted (light cyan) and proofreading (light grey) UDP-GlcA-bound CvHAS structures superimposed. B, Structures of C. albicans CHS-2 (PDB ID: 7stm, light purple) and P. sojae CHS-1 (PDB ID: 7wjn, light green) bound to UDP-GlcNAc superimposed. 7stm corresponds to the proposed ‘inserted’ UDP-GlcNAc pose for CHS.
