- Split View
-
Views
-
Cite
Cite
Ian C. Schoenhofen, David J. McNally, Jean-Robert Brisson, Susan M. Logan, Elucidation of the CMP-pseudaminic acid pathway in Helicobacter pylori: synthesis from UDP-N-acetylglucosamine by a single enzymatic reaction, Glycobiology, Volume 16, Issue 9, September 2006, Pages 8C–14C, https://doi.org/10.1093/glycob/cwl010
- Share Icon Share
Abstract
Flagellin glycosylation is a necessary modification allowing flagellar assembly, bacterial motility, colonization, and hence virulence for the gastrointestinal pathogen Helicobacter pylori [Josenhans, C., Vossebein, L., Friedrich, S., and Suerbaum, S. (2002) FEMS Microbiol. Lett., 210, 165–172; Schirm, M., Schoenhofen, I.C., Logan, S.M., Waldron, K.C., and Thibault, P. (2005) Anal. Chem., 77, 7774–7782]. A causative agent of gastric and duodenal ulcers, H. pylori, heavily modifies its flagellin with the sialic acid-like sugar 5,7-diacetamido-3,5,7,9-tetradeoxy-l-glycero-α-l-manno-nonulosonic acid (pseudaminic acid). Because this sugar is unique to bacteria, its biosynthetic pathway offers potential as a novel therapeutic target. We have identified six H. pylori enzymes, which reconstitute the complete biosynthesis of pseudaminic acid, and its nucleotide-activated form CMP-pseudaminic acid, from UDP-N-acetylglucosamine (UDP-GlcNAc). The pathway intermediates and final product were identified from monitoring sequential reactions with nuclear magnetic resonance (NMR) spectroscopy, thereby confirming the function of each biosynthetic enzyme. Remarkably, the conversion of UDP-GlcNAc to CMP-pseudaminic acid was achieved in a single reaction combining six enzymes. This represents the first complete in vitro enzymatic synthesis of a sialic acid-like sugar and sets the groundwork for future small molecule inhibitor screening and design. Moreover, this study provides a strategy for efficient large-scale synthesis of novel medically relevant bacterial sugars that has not been attainable by chemical methods alone.
Introduction
Sialic acids, namely N- and O-acyl derivatives of 5-amino-3,5-dideoxy-d-glycero-d-galacto-nonulosonic acid (neuraminic acid), are α-keto acids with a 9-carbon backbone. A major component of glycoproteins and glycolipids in eukaryotes, sialic acids, have been shown to mediate a diverse range of cell–cell and cell–molecule interactions (Vimr et al., 2004). In 1984, a novel class of nonulosonic acids was identified from the discovery of 5,7-diacetamido-3,5,7,9-tetradeoxy-l-glycero-α-l-manno-nonulosonic acid (pseudaminic acid) in the lipopolysaccharides (LPS) of Pseudomonas aeruginosa (Knirel et al., 1984). This class of sialic acid-like sugars appears to be unique to microorganisms and consists of 5,7-diacetamido-3,5,7,9-tetradeoxy-nonulosonate derivatives that can exhibit configurational differences compared with neuraminic acid (Knirel et al., 2003). Since 1984, these sugars have been found in many Gram-negative bacterial species as constituents of important cell surface glycoconjugates, such as LPS (Knirel et al., 2003), capsular polysaccharide (Kiss et al., 2001), pili (Castric et al., 2001), and flagella (Thibault et al., 2001; Schirm et al., 2003), all of which are important for pathogenesis. The reason bacteria decorate their cell surface with these sialic acid-like sugars is unknown, but these modifications could be expected to influence pathogenesis through bacterial adhesion, invasion, and immune evasion (Hsu et al., 2006).
Unlike sialic acids, little is known about the biosynthetic steps leading to 5,7-diacetamido-3,5,7,9-tetradeoxy-nonulosonate sugars in bacteria. Based on neuraminic acid biosynthesis, these sugars are likely synthesized via condensation of a 6-carbon unit with the 3-carbon molecule pyruvate. Chou and others (2005) recently demonstrated this by the chemo-enzymatic synthesis of pseudaminic acid using a synthetically derived form of 2,4-diacetamido-2,4,6-trideoxy-β-l-altropyranose, pyruvate, and a neuraminic acid synthase homolog from Campylobacter jejuni. However, the detailed biosynthetic steps leading to the 6-carbon precursor, which largely confers the stereochemical and functional nature of the respective nonulosonate (Schoenhofen, Lunin, et al., 2006), are ill-defined. The importance of these nonulosonates in virulence-associated cell surface structures of bacterial pathogens, such as the flagella of Helicobacter pylori—the only bacterium to be associated with cancer (Blaser, 1992; Megravd, 2005), prompted us to completely define the CMP-pseudaminic acid biosynthetic pathway. The elucidation of this bacterial nonulosonate pathway, and others, will have implications in the development of therapeutics to treat the ever-expanding list of antibiotic-resistant bacterial species.
Results and discussion
The identification of H. pylori enzymes responsible for the biosynthesis of CMP-pseudaminic acid (VII) involved a “systems-based” approach including genomic, metabolomic, and functional analyses. Recently, we identified and characterized the initial two steps from UDP-N-acetyl-glucosamine (UDP-GlcNAc) (I), involving the NAD(P)+-dependent dehydratase/epimerase PseB and the pyridoxal phosphate (PLP)-dependent aminotransferase PseC (Schoenhofen, Lunin, et al., 2006; Schoenhofen, McNally, et al., 2006). As previously mentioned, Chou and others (2005) identified the pseudaminic acid synthase from C. jejuni, responsible for the condensation of pyruvate with 2,4-diacetamido-2,4,6-trideoxy-β-l-altropyranose (V), forming pseudaminic acid (VI). The H. pylori homolog HP0178, namely PseI in agreement with recent gene nomenclature (Guerry et al., 2006), was identified through sequence comparisons with this synthase and that for neuraminic acid (Schirm et al., 2003). In an attempt to identify the remaining components necessary for the conversion of I to VII, we performed gene comparisons using bacterial species known to contain VI. These included H. pylori 26695 (Schirm et al., 2003) and C. jejuni 81–176 (Thibault et al., 2001), as well as genes from the symbiotic mega-plasmid pRm41c of Sinorhizobium meliloti Rm41 (Kiss et al., 2001), a plasmid responsible for the production of a capsule containing VI. This analysis highlighted PseF (HP0326A), PseG (HP0326B), and PseH (HP0327) as potential candidates. The phenotypic and metabolomic analyses of the respective isogenic knockout mutants, in either H. pylori or C. jejuni (Schirm et al., 2003; Guerry et al., 2006), further confirmed their involvement in the CMP-pseudaminic acid biosynthetic pathway. Through protein database sequence comparisons and insights gained from the chemo-enzymatic synthesis of VI (Chou et al., 2005), we screened PseF, PseG, PseH, and PseI for CMP-pseudaminic acid synthetase, nucleotidase, N-acetyltransferase, and pseudaminic acid synthase activity, respectively.
H. pylori PseF, PseG, PseH, and PseI, as well as C. jejuni PseG and PseH, were expressed in Escherichia coli and purified to near homogeneity (Figure 1) by a single nickel affinity chromatography step. When needed, PseB and PseC enzymes were purified as previously reported (Schoenhofen, McNally, et al., 2006). Monitoring the enzymatic reactions directly in a nuclear magnetic resonance (NMR) tube enabled the complete assignment of intermediates within the CMP-pseudaminic acid pathway (Table I and Supplementary Figures 1–3; NMR data for I, II, III, and V are reported elsewhere [Chou et al., 2005; Schoenhofen, McNally, et al., 2006]). Sugars implicated in this pathway were readily identified with NMR because the signals originating from anomeric and deoxy (CH3) protons were different for each sugar. By the sequential addition of putative biosynthetic enzymes and their expected cofactors and co-substrates (Olsen and Roderick, 2001; Tarbouriech et al., 2001; Munster-Kuhnel et al., 2004; Chou et al., 2005; Schoenhofen, Lunin, et al., 2006; Schoenhofen, McNally, et al., 2006), the entire CMP-pseudaminic acid biosynthetic pathway, starting from I, was determined and proceeds as follows: C-4,6 dehydration and C-5 epimerization of I by PseB forming UDP-2-acetamido-2,6-dideoxy-β-l-arabino-hexos-4-ulose (II), C-4 aminotransfer of II by PseC forming UDP-4-amino-4,6-dideoxy-β-l-AltNAc (III), N-4 acetylation of III by PseH forming UDP-2,4-diacetamido-2,4,6-trideoxy-β-l-altropyranose (IV), removal of UDP from C-1 of IV by PseG forming 2,4-diacetamido-2,4,6-trideoxy-β-l-altropyranose (V), condensation of V with pyruvate by PseI forming pseudaminic acid (VI), and finally activation of VI with CMP by PseF forming CMP-pseudaminic acid (VII) (Figure 2).
Compound . | 1H . | δH (ppm) . | 13C . | δC (ppm) . | JH,H (Hz) . |
---|---|---|---|---|---|
IV | H1 | 5.64 | C1 | 93.9 | J1,2 (2.3) |
JH,P (8.5) | |||||
H2 | 4.16 | C2 | 53.8 | J2,3 (4.5) | |
H3 | 4.02 | C3 | 67.9 | J3,4 (3.2) | |
H4 | 3.88 | C4 | 51.6 | J4,5 (9.3) | |
H5 | 4.02 | C5 | 71.6 | J5,6 (6.4) | |
H6 | 1.26 | C6 | 18.4 | ||
2NAc-CH3 | 2.07 | 2NAc-CH3 | 22.6 | ||
2NAc-NH | 8.17 | JNH,2 (8.0) | |||
2NAc-C=O | 175.1 | ||||
4NAc-CH3 | 2.03 | 4NAc-CH3 | 22.6 | ||
4NAc-NH | 8.04 | JNH,4 (8.7) | |||
4NAc-C=O | 175.1 | ||||
VI | C1 | 177.4 | |||
C2 | 97.3 | ||||
H3ax | 1.78 | C3 | 35.6 | J3ax,3eq (13.3) | |
J3ax,4 (12.6) | |||||
H3eq | 1.93 | C3 | 35.6 | J3eq,4 (5.0) | |
H4 | 4.17 | C4 | 66.1 | J4,5 (4.0) | |
H5 | 4.24 | C5 | 49.6 | J5,6 (1.8) | |
H6 | 4.02 | C6 | 70.8 | J6,7 (10.3) | |
H7 | 4.15 | C7 | 53.8 | J7,8 (3.5) | |
H8 | 4.11 | C8 | 67.7 | J8,9 (6.5) | |
H9 | 1.10 | C9 | 16.2 | ||
5NAc-CH3 | 2.01 | 5NAc-CH3 | 22.7 | ||
5NAc-NH | 8.57 | JNH,5 (10.0) | |||
5NAc-C=O | 175.5 | ||||
7NAc-CH3 | 1.98 | 7NAc-CH3 | 22.9 | ||
7NAc-NH | 7.95 | JNH,7 (10.1) | |||
7NAc-C=O | 174.7 | ||||
VII | C1 | 174.7 | |||
C2 | 100.6 | ||||
H3ax | 1.60 | C3 | 36.7 | J3ax,3eq (13.5) | |
J3ax,4 (12.0) | |||||
H3eq | 2.22 | C3 | 36.7 | J3eq,4 (4.6) | |
H4 | 4.23 | C4 | 65.5 | J4,5 (3.9) | |
H5 | 4.28 | C5 | 49.5 | J5,6 (1.4) | |
H6 | 4.30 | C6 | 73.4 | J6,7 (9.9) | |
H7 | 4.02 | C7 | 54.4 | J7,8 (4.9) | |
H8 | 4.11 | C8 | 69.4 | J8,9 (6.5) | |
H9 | 1.20 | C9 | 17.9 | ||
5NAc-CH3 | 2.00 | 5NAc-CH3 | 22.6 | ||
5NAc-NH | 8.50 | JNH,5 (9.7) | |||
5NAc-C=O | 175.2 | ||||
7NAc-CH3 | 1.97 | 7NAc-CH3 | 22.7 | ||
7NAc-NH | 8.06 | JNH,7 (10.4) | |||
7NAc-C=O | 174.0 |
Compound . | 1H . | δH (ppm) . | 13C . | δC (ppm) . | JH,H (Hz) . |
---|---|---|---|---|---|
IV | H1 | 5.64 | C1 | 93.9 | J1,2 (2.3) |
JH,P (8.5) | |||||
H2 | 4.16 | C2 | 53.8 | J2,3 (4.5) | |
H3 | 4.02 | C3 | 67.9 | J3,4 (3.2) | |
H4 | 3.88 | C4 | 51.6 | J4,5 (9.3) | |
H5 | 4.02 | C5 | 71.6 | J5,6 (6.4) | |
H6 | 1.26 | C6 | 18.4 | ||
2NAc-CH3 | 2.07 | 2NAc-CH3 | 22.6 | ||
2NAc-NH | 8.17 | JNH,2 (8.0) | |||
2NAc-C=O | 175.1 | ||||
4NAc-CH3 | 2.03 | 4NAc-CH3 | 22.6 | ||
4NAc-NH | 8.04 | JNH,4 (8.7) | |||
4NAc-C=O | 175.1 | ||||
VI | C1 | 177.4 | |||
C2 | 97.3 | ||||
H3ax | 1.78 | C3 | 35.6 | J3ax,3eq (13.3) | |
J3ax,4 (12.6) | |||||
H3eq | 1.93 | C3 | 35.6 | J3eq,4 (5.0) | |
H4 | 4.17 | C4 | 66.1 | J4,5 (4.0) | |
H5 | 4.24 | C5 | 49.6 | J5,6 (1.8) | |
H6 | 4.02 | C6 | 70.8 | J6,7 (10.3) | |
H7 | 4.15 | C7 | 53.8 | J7,8 (3.5) | |
H8 | 4.11 | C8 | 67.7 | J8,9 (6.5) | |
H9 | 1.10 | C9 | 16.2 | ||
5NAc-CH3 | 2.01 | 5NAc-CH3 | 22.7 | ||
5NAc-NH | 8.57 | JNH,5 (10.0) | |||
5NAc-C=O | 175.5 | ||||
7NAc-CH3 | 1.98 | 7NAc-CH3 | 22.9 | ||
7NAc-NH | 7.95 | JNH,7 (10.1) | |||
7NAc-C=O | 174.7 | ||||
VII | C1 | 174.7 | |||
C2 | 100.6 | ||||
H3ax | 1.60 | C3 | 36.7 | J3ax,3eq (13.5) | |
J3ax,4 (12.0) | |||||
H3eq | 2.22 | C3 | 36.7 | J3eq,4 (4.6) | |
H4 | 4.23 | C4 | 65.5 | J4,5 (3.9) | |
H5 | 4.28 | C5 | 49.5 | J5,6 (1.4) | |
H6 | 4.30 | C6 | 73.4 | J6,7 (9.9) | |
H7 | 4.02 | C7 | 54.4 | J7,8 (4.9) | |
H8 | 4.11 | C8 | 69.4 | J8,9 (6.5) | |
H9 | 1.20 | C9 | 17.9 | ||
5NAc-CH3 | 2.00 | 5NAc-CH3 | 22.6 | ||
5NAc-NH | 8.50 | JNH,5 (9.7) | |||
5NAc-C=O | 175.2 | ||||
7NAc-CH3 | 1.97 | 7NAc-CH3 | 22.7 | ||
7NAc-NH | 8.06 | JNH,7 (10.4) | |||
7NAc-C=O | 174.0 |
Compound . | 1H . | δH (ppm) . | 13C . | δC (ppm) . | JH,H (Hz) . |
---|---|---|---|---|---|
IV | H1 | 5.64 | C1 | 93.9 | J1,2 (2.3) |
JH,P (8.5) | |||||
H2 | 4.16 | C2 | 53.8 | J2,3 (4.5) | |
H3 | 4.02 | C3 | 67.9 | J3,4 (3.2) | |
H4 | 3.88 | C4 | 51.6 | J4,5 (9.3) | |
H5 | 4.02 | C5 | 71.6 | J5,6 (6.4) | |
H6 | 1.26 | C6 | 18.4 | ||
2NAc-CH3 | 2.07 | 2NAc-CH3 | 22.6 | ||
2NAc-NH | 8.17 | JNH,2 (8.0) | |||
2NAc-C=O | 175.1 | ||||
4NAc-CH3 | 2.03 | 4NAc-CH3 | 22.6 | ||
4NAc-NH | 8.04 | JNH,4 (8.7) | |||
4NAc-C=O | 175.1 | ||||
VI | C1 | 177.4 | |||
C2 | 97.3 | ||||
H3ax | 1.78 | C3 | 35.6 | J3ax,3eq (13.3) | |
J3ax,4 (12.6) | |||||
H3eq | 1.93 | C3 | 35.6 | J3eq,4 (5.0) | |
H4 | 4.17 | C4 | 66.1 | J4,5 (4.0) | |
H5 | 4.24 | C5 | 49.6 | J5,6 (1.8) | |
H6 | 4.02 | C6 | 70.8 | J6,7 (10.3) | |
H7 | 4.15 | C7 | 53.8 | J7,8 (3.5) | |
H8 | 4.11 | C8 | 67.7 | J8,9 (6.5) | |
H9 | 1.10 | C9 | 16.2 | ||
5NAc-CH3 | 2.01 | 5NAc-CH3 | 22.7 | ||
5NAc-NH | 8.57 | JNH,5 (10.0) | |||
5NAc-C=O | 175.5 | ||||
7NAc-CH3 | 1.98 | 7NAc-CH3 | 22.9 | ||
7NAc-NH | 7.95 | JNH,7 (10.1) | |||
7NAc-C=O | 174.7 | ||||
VII | C1 | 174.7 | |||
C2 | 100.6 | ||||
H3ax | 1.60 | C3 | 36.7 | J3ax,3eq (13.5) | |
J3ax,4 (12.0) | |||||
H3eq | 2.22 | C3 | 36.7 | J3eq,4 (4.6) | |
H4 | 4.23 | C4 | 65.5 | J4,5 (3.9) | |
H5 | 4.28 | C5 | 49.5 | J5,6 (1.4) | |
H6 | 4.30 | C6 | 73.4 | J6,7 (9.9) | |
H7 | 4.02 | C7 | 54.4 | J7,8 (4.9) | |
H8 | 4.11 | C8 | 69.4 | J8,9 (6.5) | |
H9 | 1.20 | C9 | 17.9 | ||
5NAc-CH3 | 2.00 | 5NAc-CH3 | 22.6 | ||
5NAc-NH | 8.50 | JNH,5 (9.7) | |||
5NAc-C=O | 175.2 | ||||
7NAc-CH3 | 1.97 | 7NAc-CH3 | 22.7 | ||
7NAc-NH | 8.06 | JNH,7 (10.4) | |||
7NAc-C=O | 174.0 |
Compound . | 1H . | δH (ppm) . | 13C . | δC (ppm) . | JH,H (Hz) . |
---|---|---|---|---|---|
IV | H1 | 5.64 | C1 | 93.9 | J1,2 (2.3) |
JH,P (8.5) | |||||
H2 | 4.16 | C2 | 53.8 | J2,3 (4.5) | |
H3 | 4.02 | C3 | 67.9 | J3,4 (3.2) | |
H4 | 3.88 | C4 | 51.6 | J4,5 (9.3) | |
H5 | 4.02 | C5 | 71.6 | J5,6 (6.4) | |
H6 | 1.26 | C6 | 18.4 | ||
2NAc-CH3 | 2.07 | 2NAc-CH3 | 22.6 | ||
2NAc-NH | 8.17 | JNH,2 (8.0) | |||
2NAc-C=O | 175.1 | ||||
4NAc-CH3 | 2.03 | 4NAc-CH3 | 22.6 | ||
4NAc-NH | 8.04 | JNH,4 (8.7) | |||
4NAc-C=O | 175.1 | ||||
VI | C1 | 177.4 | |||
C2 | 97.3 | ||||
H3ax | 1.78 | C3 | 35.6 | J3ax,3eq (13.3) | |
J3ax,4 (12.6) | |||||
H3eq | 1.93 | C3 | 35.6 | J3eq,4 (5.0) | |
H4 | 4.17 | C4 | 66.1 | J4,5 (4.0) | |
H5 | 4.24 | C5 | 49.6 | J5,6 (1.8) | |
H6 | 4.02 | C6 | 70.8 | J6,7 (10.3) | |
H7 | 4.15 | C7 | 53.8 | J7,8 (3.5) | |
H8 | 4.11 | C8 | 67.7 | J8,9 (6.5) | |
H9 | 1.10 | C9 | 16.2 | ||
5NAc-CH3 | 2.01 | 5NAc-CH3 | 22.7 | ||
5NAc-NH | 8.57 | JNH,5 (10.0) | |||
5NAc-C=O | 175.5 | ||||
7NAc-CH3 | 1.98 | 7NAc-CH3 | 22.9 | ||
7NAc-NH | 7.95 | JNH,7 (10.1) | |||
7NAc-C=O | 174.7 | ||||
VII | C1 | 174.7 | |||
C2 | 100.6 | ||||
H3ax | 1.60 | C3 | 36.7 | J3ax,3eq (13.5) | |
J3ax,4 (12.0) | |||||
H3eq | 2.22 | C3 | 36.7 | J3eq,4 (4.6) | |
H4 | 4.23 | C4 | 65.5 | J4,5 (3.9) | |
H5 | 4.28 | C5 | 49.5 | J5,6 (1.4) | |
H6 | 4.30 | C6 | 73.4 | J6,7 (9.9) | |
H7 | 4.02 | C7 | 54.4 | J7,8 (4.9) | |
H8 | 4.11 | C8 | 69.4 | J8,9 (6.5) | |
H9 | 1.20 | C9 | 17.9 | ||
5NAc-CH3 | 2.00 | 5NAc-CH3 | 22.6 | ||
5NAc-NH | 8.50 | JNH,5 (9.7) | |||
5NAc-C=O | 175.2 | ||||
7NAc-CH3 | 1.97 | 7NAc-CH3 | 22.7 | ||
7NAc-NH | 8.06 | JNH,7 (10.4) | |||
7NAc-C=O | 174.0 |
For efficient enzymatic synthesis of VII, a few reaction optimizations were required. When all six CMP-pseudaminic acid biosynthetic enzymes were combined in a reaction with I, synthesis of VII was successful, although we observed a significant reduction in product yield as compared with those obtained with the first three, four, or five biosynthetic pathway enzymes (Figure 3A). In addition to low yields of VII, the inclusion of PseF in the “one-pot” reaction also affected the production of VI, indicative of feedback inhibition in the pathway. A similar regulatory effect was observed in the CMP-sialic acid pathway, in which CMP-sialic acid was found to be an allosteric inhibitor of an early pathway enzyme (Munster-Kuhnel et al., 2004). However, by combining the first five enzymes in a “one-pot” reaction, we were able to obtain 100% conversion of I to VI after 20 h, a yield of pseudaminic acid which surpasses that obtained using chemical methods alone. Chemical methods using 2,4-diacetamido-2,4,6-trideoxy-β-l-allose and oxaloacetic acid yielded only 3% conversion of the 2,4-diacetamido-2,4,6-trideoxyhexose to pseudaminic acid (Knirel et al., 2003). Interestingly, when NMR was used to monitor the “one-pot” reaction containing the first five enzymes, only signals from I and VI were observed, indicating the efficiency and speed of the coupled enzymatic reaction. As well, our characterization of the enzyme-derived product VI (Table I and Supplementary Figure 2) confirms the structural assignment reported for that derived from chemo-enzymatic synthesis (Chou et al., 2005). To further optimize for the efficient production of VII, we used the enzymatically derived product VI and incubated this with PseF at either pH 7.3 or pH 9.0 (Figure 3B and C), in the presence or absence of Mg2+ (data not shown). As expected from the conditions used for CMP-sialic acid synthetase activity (Karwaski et al., 2002), alkaline pH and Mg2+ were required for optimal PseF activity. Using a two-step enzymatic process, where the initially generated product VI was CMP activated using PseF, we obtained 100% conversion of I to VII producing milligram yields before scale-up. In summary, we have outlined a facile strategy for the efficient enzymatic production of bacterial nonulosonates using the commercially available substrate UDP-GlcNAc. By scaling-up this process, gram-scale synthesis of these sugars and their intermediates can be achieved easily. For this, one could include cofactor regeneration (Koeller and Wong, 2001), such as that for cytidine triphosphate (CTP) and acetyl-coenzymeA (acetyl-CoA), thus making the process economically and industrially feasible.
The elucidation of bacterial nonulosonate pathways is important for several reasons. One is that these pathways may be exploited for the generation of novel antibiotics. Presently, there is great interest in the manipulation of deoxy-sugar biosynthetic pathways for the purposes of engineering novel antitumor and antibiotic drugs (Chen et al., 2000; Rodriguez et al., 2002; Nedal and Zotchev, 2004; Griffith et al., 2005). The macrolide antibiotic erythromycin, antifungal polyene amphotericin B, parasiticide avermectin, and anticancer agent doxorubicin all possess 6-deoxy-hexose sugars that are responsible for novel pharmacological properties. Because 6-deoxy-hexose metabolism is a feature inherent to bacterial nonulosonate pathways, there is tremendous glycoengineering potential, in that these biosynthetic steps may now be manipulated for the preparation of novel therapeutic compounds. Adding to this potential is the flexibility demonstrated by a similar pathway, that of sialic acid, in accepting non-natural metabolic intermediates (Sampathkumar et al., 2006).
Furthermore, because these sialic acid-like sugars are found on virulence determinants of important pathogens and are unique to microorganisms, their pathways can be the targets of therapeutic agents. As the modes of action of current antibacterials are presently limited, targeting either the production of folic acid, nucleic acids, proteins or the cell wall, new strategies to combat emerging drug-resistant species are urgently required. The reliance of H. pylori pathogenicity on pseudaminic acid synthesis provides an attractive therapeutic alternative. Conveniently, the entire pseudaminic acid pathway can now be screened for inhibitors in vitro, rather than just specific enzymes. This can be accomplished by combining all five of the biosynthetic enzymes, UDP-GlcNAc, and potential inhibitors in a microtiter plate format and screening for phosphate release, which only occurs during the last biosynthetic step. A similar strategy, monitoring the depletion of nicotinamide adenine dinucleotide phosphate (NADPH), was employed to screen for inhibitors of dTDP-glucose to dTDP-rhamnose conversion (Ma et al., 2001). In addition, these sugars can be used as vaccine candidates because the glycan found on P. aeruginosa pili, containing VI, is recognized by host antibodies (Castric et al., 2001; Comer et al., 2002). These antibodies also recognize the LPS of P. aeruginosa, containing VI, and inhibit twitching motility.
Finally, in focusing on H. pylori and C. jejuni flagellin glycosylation, our ability to produce CMP-pseudaminic acid in large quantities permits the in vitro characterization of the hitherto elusive glycosyltransferases. As well, the biosynthesis of N-5,7- and O-8-substituted pseudaminic acid derivatives, such as the N-acetamidino and O-acetyl forms found in C. jejuni (Schirm et al., 2005), may be unraveled.
Materials and methods
His6-tagged protein expression and purification
DNA techniques and plasmid construction were similar to previously described methods (Schoenhofen, McNally, et al., 2006). Vector or recombinant plasmids were transformed by electroporation into electrocompetent Top10F’ or DH10B (Invitrogen, Burlington, Ontario, Canada) E. coli cells for cloning purposes or BL21[DE3] (Novagen, Madison, WI) E. coli cells for protein production, except for the expression clone pNRC36.3 that was electroporated into BL21-CodonPlus[DE3]-RIL (Novagen) E. coli cells. Polymerase chain reaction (PCR) was used to amplify H. pylori 26695 DNA or C. jejuni 11168 DNA for subsequent cloning. A list of cloning vectors and recombinant plasmids is provided in Supplementary Table I, and pertinent oligonucleotides are provided in Supplementary Table II. pET30a and pFO4 recombinant plasmids were sequenced using both forward and reverse T7 primers, as well as NRC175 and NRC160, respectively. Newly constructed plasmids are pNRC129.2, encoding an N-terminal His6-tagged derivative of HP0327 or H. pylori PseH; pNRC128.1, encoding an N-terminal His6-tagged derivative of Cj1313 or C. jejuni PseH; pNRC131.1, encoding a C-terminal His6-tagged derivative of HP0326B or H. pylori PseG; pNRC17.1, encoding a C-terminal His6-tagged derivative of Cj1312 or C. jejuni PseG; pNRC36.3, encoding an N-terminal His6-tagged derivative of HP0178 or H. pylori PseI; and pNRC38.1, encoding an N-terminal His6-tagged derivative of HP0326A or H. pylori PseF.
Typically, each expression strain was grown in 1–4 L of 2 × yeast tryptone (Schoenhofen, McNally, et al., 2006), depending on expression level, with kanamycin (50 µg mL–1), ampicillin (75 µg mL–1), or ampicillin and chloramphenicol (100 and 40 µg mL–1) for selection. The cultures were grown at 30°C, induced at an OD600 of 0.6 with 0.1 mM isopropyl-1-thio-β-d-galactopyranoside, and harvested 2.75 h later. Cell pellets were resuspended in lysis buffer (50 mM sodium phosphate, pH 7.3, 400 mM NaCl, 10 mM β-mercaptoethanol) containing 10 mM imidazole and complete protease inhibitor mixture, EDTA-free (Roche Applied Science, Laval, Quebec, Canada). After the addition of 10 µg mL–1 of RNaseA and DNaseI (Roche Applied Science), the cells were disrupted by two passes through an emulsiflex C5 (20,000 psi). Lysates were centrifuged at 100,000 × g for 50 min at 4°C, and the supernatant fraction was applied to a 3-mL nickel–nitrilotriacetic acid (Qiagen, Mississavga, Ontario, Canada) column using a flow rate of 1 mL min–1. After sample application, the column was washed with 10 column volumes of 10 mM imidazole lysis buffer. To elute the protein of interest, we applied a linear gradient from 10 to 100 mM imidazole, in lysis buffer, over 30 column volumes to the column before a final pulse of 10 column volumes of 200 mM imidazole lysis buffer. Fractions containing the purified protein of interest, as determined by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) (12.5%) and Coomassie staining, were pooled and dialyzed against dialysis buffer (25 mM sodium phosphate, pH 7.3, 50 mM NaCl) overnight at 4°C. Lysis and dialysis buffers for the purification of PseGHis6 and His6PseF proteins additionally contained 1 mM MgCl2. Protein concentration was measured spectrophotometrically using A280 0.1% values (H. pylori proteins: PseBHis6, 0.536; His6PseC, 0.386; His6PseH, 1.037; PseGHis6, 0.706; His6PseI, 0.513; His6PseF, 0.475; and C. jejuni proteins: PseBHis6, 0.669; His6PseC, 0.635; His6PseH, 0.878; PseGHis6, 0.740). Yields of purified protein were typically 7–11 mg L–1 of cell culture.
Enzymatic reactions and metabolite purification
Enzymatic reactions were performed in dialysis buffer at 37°C with ∼85 µg mL–1 respective protein concentration using chemicals from Sigma, Oakville, Ontario, Canada (unless otherwise indicated). The stepwise enzymatic synthesis of intermediates or products (III→IV→V→VI→VII) was accomplished using four separate 4-h enzymatic reactions containing 1 mM UDP-GlcNAc, 1 mM PLP, 10 mM l-Glu, H. pylori PseBHis6 and His6PseC, as well as 1.5 mM acetyl-CoA, 1.5 mM phosphoenolpyruvate (PEP), 1.5 mM CTP, H. pylori His6PseH, PseGHis6, His6PseI, and His6PseF as appropriate. Activities of C. jejuni His6PseH and PseGHis6 were tested similarly. Optimization of the H. pylori PseF reaction involved monitoring the reaction kinetics by 1H NMR (see NMR spectroscopy) using 0.5 mM pseudaminic acid, 130 µg mL–1 His6PseF, 1.5 mM CTP, and 10 mM MgCl2 at either pH 7.3 or pH 9.0, 25°C.
Large-scale enzymatic synthesis of IV was accomplished using a 60-mL reaction containing 1 mM UDP-GlcNAc, 1 mM PLP, 10 mM l-Glu, 1.75 mM acetyl-CoA, and ∼23 mg each of C. jejuni PseBHis6, His6PseC, and His6PseH; that of VI was accomplished using a 27-mL reaction containing 0.75 mM of preparation IV above, 1.5 mM PEP, as well as H. pylori PseGHis6 and His6PseI; and that of VII was accomplished using a 15-mL reaction containing 0.5 mM of preparation VI above, 1.5 mM CTP, 10 mM MgCl2, as well as 180 µg mL–1H. pylori His6PseF. After passage through an Amicon Ultra-15 (10,000 molecular weight cut-off) filter, large-scale sugar preparations were lyophilized and desalted/purified using a Bio-Gel P-2 (Bio-Rad, Mississavga, Ontario, Canada) column in 25 mM ammonium bicarbonate, pH 7.9. Quantification of the nucleotide-linked sugar preparations was determined using the molar extinction coefficients of UDP (ε260 = 10,000) and CMP (ε260 = 7400).
NMR spectroscopy
Enzymatic reactions were carried out in 3-mm NMR tubes at 25°C in 90% aqueous buffer (10% D2O, 25 mM NaPO4, 50 mM NaCl, pH 7.3) and were monitored directly with NMR spectroscopy through the acquisition of 1H spectrum at regular time intervals (i.e., 10 min) using a Varian Inova 500 MHz (1H) spectrometer with a Varian Z-gradient 3-mm probe. To observe major and minor enzymatic products, we increased the signal-to-noise ratio by acquiring 128 scans per time point. To characterize the structures of compounds IV–VII, we exchanged purified material into 100% D2O. Standard pulse sequences from Varian and selective one-dimensional total correlation spectroscopy (TOCSY) and nuclear Overhauser effect spectroscopy (NOESY) experiments were used for complete residue assignments, measurement of coupling constants (JH,H), and nuclear Overhauser effects (Schoenhofen, McNally, et al., 2006). NMR experiments were typically performed at 25°C with suppression of the H2O or deuterated deuterated H2O resonance at δH 4.78 ppm. For proton and carbon experiments, the methyl resonance of acetone was used as an internal reference (δH 2.225 ppm and δC 31.07 ppm). For Figure 3A, relative yields were determined by comparing the integrals calculated for the anomeric proton signals of I, to those for the C-6 CH3 protons of IV and V, and to the C-9 CH3 protons of VI and VII. For Figure 3B and C, integrals were calculated for the C-9 CH3 proton signals of VI and VII.
Supplementary data
Supplementary data are available at Glycobiology online (http://glycob.oxfordjournals.org/).
Conflict of interest statement
None declared.
aminotransferase; PseH, N-acetyltransferase; PseG, nucleotidase; PseI, pseudaminic acid synthase; PseF, CMP-pseudaminic acid synthetase; and (I) UDP-GlcNAc; (II) UDP-2-acetamido-2,6-dideoxy-b-l-arabino-hexos-4-ulose; (III) UDP-4-amino-4,6-dideoxy-b-l-AltNAc; (IV) UDP-2,4-diacetamido-2,4,6-trideoxy-b-l-altropyranose; (V) 2,4-diacetamido-2,4,6-trideoxy-b-l-altropyranose; (VI) pseudaminic acid; (VII) CMP-pseudaminic acid. The assignment of roman numerals to each compound is consistent with label designations found throughout the text. Pyranose rings are shown as their predominant chair conformation in solution determined from nuclear Overhauser effects (NOEs) and JH,H coupling constants.
Acknowledgments
We thank G. Selkirk and A. Aubry for technical assistance with initial pseF and pseG cloning and N. Khieu for maintenance of NMR instruments. We are grateful to M. Cygler, A. Matte, and D. Whitfield for helpful discussions and comments. This study was supported by the National Research Council Genomics and Health Initiative.
References