TY - JOUR
AU - Eggeling, Lothar
AB - Abstract The cell wall mycolyl-arabinogalactan (AG)–peptidoglycan complex is essential in mycobacterial species, such as Mycobacterium tuberculosis, and is the target of several antitubercular drugs. For instance, ethambutol (EMB) targets AG biosynthesis through inhibition of the arabinofuranosyltransferases Mt-EmbA and Mt-EmbB, as well as the single Emb from Corynebacterium glutamicum. Here, we present for the first time an experimental analysis of the membrane topology of Emb. The domain organization clearly positions highly conserved loop regions, like the recognized glycosyltransferase C motif and the hydrophilic C-terminus towards the periplasmic side of the cell. Moreover, the assignment and orientation of hydrophobic segments identified a loop region, which might dip into the membrane and could possibly line a transportation channel for the emerging substrate. Site-directed mutations introduced into plasmid-encoded Cg-emb were analyzed in a C. glutamicumΔemb strain for their AG glycosyl composition and linkage analysis. Mutations analyzed did not perturb galactan synthesis; however, D297A produced a dramatically reduced arabinan content and prevented growth, indicating an inactive Emb. A second D298A mutation also drastically reduced arabinan content; however, growth of the corresponding mutant was not altered, indicating a certain tolerance of this mutation in terms of Emb function. A W659L–P667A–Q674E triple mutation in the chain length regulation motif (Pro-motif) resulted in a reduced arabinose deposition in AG but retained all arabinofuranosyl linkages. Taken together, the data clearly define important residues of Emb involved in arabinan domain formation and, for the first time, shed new light on the topology of this important enzyme. glycosyltransferase, arabinosyltransferase, arabinogalactan, cell wall, ethambutol Introduction The Corynebacterianeae represent a distinct group within Gram-positive bacteria, with prominent members being the human pathogens Mycobacterium tuberculosis, M. leprae, Corynebacterium diphtheriae, and C. jeikeium (Bloom and Murray 1992). In addition, nonpathogenic bacteria belong to this taxon, such as C. glutamicum, C. efficiens, and C. ammoniagenes which are used in the industrial production of amino acids and nucleotides (Eggeling and Bott 2005). A common feature of the Corynebacterianeae is that they possess an unusual cell wall architecture (McNeil et al. 1990, 1991; Besra et al. 1995). The cell wall is dominated by mycolic acids and an essential heteropolysaccharide arabinogalactan (AG) linked to peptidoglycan via a disaccharide linker unit, l-Rhap-(1 → 4)-α-d-GlcNAc-phosphate, thus forming the mycolyl-AG–peptidoglycan (mAGP) complex (McNeil et al. 1990, 1991; Besra et al. 1995; Alderwick et al. 2005). The AG is composed of approximately 30 galactofuranosyl residues of alternating β(1 → 6) and β(1 → 5)-Galf linkages (Mikusova et al., 2000; Kremer et al. 2001). The galactan domain is further glycosylated at the 8th, 10th, and 12th residue by arabinan motifs, made up primarily of α(1 → 5), α(1 → 3), and β(1 → 2)-Araf residues (Alderwick et al. 2005), yielding arabinan domains of approximately 25 residues in size (Besra et al. 1995). The antituberculosis drug ethambutol (EMB) was shown to specifically inhibit AG biosynthesis (Takayama and Kilburn 1989). The precise molecular target of EMB occupies the embCAB locus in M. tuberculosis and its encoded proteins (Telenti et al. 1997). To further define the role of the EmbCAB proteins in cell wall arabinan biosynthesis, the embC, embA, and embB genes were disrupted individually in M. smegmatis, however, with the sequences in the genome retained (Escuyer et al. 2001; Zhang et al. 2003). Although all three mutants were viable, only the crucial terminal Ara6 motif, which is the template for mycolylation in AG (McNeil et al. 1991), was altered in both Ms-embA and Ms-embB mutants (Escuyer et al. 2001). This suggested that EmbA and EmbB are involved in the formation of the terminal Ara6 motif in AG and also presumably compensated for each other in the Ms-embA and Ms-embB mutants, whereas Ms-embC was involved in the formation of arabinan domains of lipoarabinomannan (LAM) (Zhang et al. 2003). Attempts to obtain deletion mutants of embA and embB in M. tuberculosis and embAB in M. smegmatis have proved unsuccessful (GSB, unpublished results), thus preventing the analysis of a simple defined sytem to further unravel the apparent complex Emb functions in Mycobacterium. In this respect, Corynebacterium glutamicum has proved useful, as it represents one of the simplest Corynebacterianeae with respect to cell wall structure and genomic organization. It possesses the basic cell wall components and lipids characteristic of this peculiar group of bacteria, and often a limited number of paralogous genes, when compared with Mycobacterium species, are present (Kalinowski et al. 2003). Moreover, it is more tolerable with respect to deletion of essential genes. In contrast to M. tuberculosis, deletion of the single Cg-emb ortholog was possible and chemical analysis of the cell wall revealed the entire absence of the arabinan domain in AG except terminal t-Araf residues, thus representing a novel truncated AG structure (Alderwick et al. 2005). More recent studies have led to the identification and characterization of the novel arabinofuranosyltransferase AftA in C. glutamicum conserved in the Corynebacterianeae, including M. tuberculosis, which catalyzes the addition of the first key arabinofuranosyl residue from the sugar donor β-d-arabinofuranosyl-l-monophosphoryl-decaprenol to the galactan domain of the cell wall, thus “priming” the galactan for further elaboration by Cg-Emb and Mt-EmbA/B, respectively (Alderwick, Seidel et al. 2006). Despite the importance of the arabinofuranosyltransferases Emb and AftA, there is little information on the structure and mechanism of these membrane-located enzymes. Overall, the Emb proteins of Corynebacterianeae are similar in size, 1094–1146 aa residues, and predictions suggest a very similar topology with transmembrane spanning helices dominating the N-terminal domain followed by a large hydrophilic C-terminal domain, probably directed towards the periplasmic side. Interestingly, in clinical isolates of EMB-resistant M. tuberculosis strains, mutations are found within embC, embA, and embB, with prominent mutations in the putative membrane embedded N-terminal part of EmbB, specifically Met 306 (Ramaswamy et al. 2000). However, mutational studies to date have focussed on Ms-EmbC, which has been shown to be involved in LAM biosynthesis (Zhang et al. 2003). Interestingly, in this regard, when point mutations within a motif characteristic for the glycosyltransferase C (GT-C) family (Liu and Mushegian 2003) were introduced into a complementing plasmid in the Ms-embC mutant, LAM synthesis was completely abolished (Berg et al. 2005). The GT-C motif is predicted to face the periplasmic side of the membrane, as does a second chain length regulation proline motif (Pro-motif) (Berg et al. 2005). In addition, point mutations within the Pro-motif when introduced into the complementing plasmid in the Ms-embC mutant led to a truncation in LAM. More recently, truncated embC alleles within the hydrophilic extracytoplasmic C-terminal domain when introduced into the complementing plasmid in the Ms-embC mutant have afforded highly truncated LAMs (Shi et al. 2006). It is clear that there is a need to deepen our knowledge on the structure and function of arabinofuranosyltransferases within the Corynebacterianeae. In the present study, we have used the C. glutamicum mutant deleted of the single emb ortholog as it provides a strong phenotypic background in terms of arabinan biosynthesis to perform a topological analysis of Cg-Emb together with a mutational study of this protein to clarify the role of emb in AG biosynthesis for the first time. Results Reporter fusions to Cg-emb A clear advantage of our studies is that C. glutamicum has only one single arabinofuranosyltransferase, with identities to the Emb proteins of M. tuberculosis, and which is proved to be involved in AG biosynthesis (Alderwick et al. 2005), whereas M. tuberculosis has three Emb proteins (Telenti et al. 1997). They all share a high degree of sequence similarity among each other. Interestingly, Cg-Emb exhibits the highest similarity of 58% to Mt-EmbC, thought to be involved in LAM biosynthesis (Zhang et al. 2003), which is remarkable. We therefore compared the wild type of C. glutamicum and its emb deletion mutant C. glutamicumΔemb in terms of their lipoglycan content. Corynebacterium glutamicum has lipomannan (Gibson et al. 2003) and a very specific LAM. Gas chromatography (GC) and GC/mass spectrometry (MS) analyses showed that LAM of C. glutamicum possesses a linear mannan backbone with only single t-Araf residue attached and this structure is retained in C. glutamicumΔemb (data not shown), whereas LAM of M. tuberculosis has a branched mannan core with chains of Araf residues attached (Chatterjee et al. 1992). This confirms that the single emb of C. glutamicum contributes to the formation of the arabinan domain in AG (Alderwick, Dover et al. 2006; Alderwick, Seidel et al. 2006), but does not affect LAM synthesis. Because of the importance of Emb proteins in terms of drug targeting and resistance to EMB (Takayama and Kilburn 1989) as well as their importance for mAGP biosynthesis, supportive information on their structure and activity is urgently needed. To this end, we fused Cg-emb in pMS3 with an alkaline phosphatase-β-galactosidase reporter cassette (Alexeyev and Winkler 1999). This system enables to localize the fusion point to the periplasmic side when alkaline phosphatase is active and β-galactosidase is inactive or to the cytoplasmic side when the enzyme activities are reversed. Using introduced restriction sites (see Materials and methods) and a set of exonuclease III treatments, Cg-emb was digested from its 3′ end, thus creating a collection of truncated Emb proteins with the fusion at their C-terminus. A total of 62 fusions were chosen conferring a blue, red, or purple color to Escherichia coli DH5α indicator plates due to an apparent alkaline phosphatase activity (Table I). Sequencing identified that these fusions cover the entire length of the polypeptide sequence from 67 to 1146 aa. The location of all fusions is given in Figure 1, with all fusion sites listed in Supplementary Table S1. Red colonies indicative of β-galactosidase activity were obtained with fusions at L471 and R585. To confirm the color appearance of the colonies, a direct enzyme assay for 16 selected clones was performed concentrating on the first two-thirds of the protein, where topology predictions were strongly deviating from each other. As shown in Table I, both phosphatase and β-galactosidase activities were well above the control and generally agreed with colony coloration. An exception was the fusion at R637, which resulted in an ambiguous purple color on plates. However, the absolute specific β-galactosidase activity, as well as the ratio of activities, directs this fusion to the cytoplasmic side. Fig. 1. Open in new tabDownload slide Topological model of the Cg-emb arabinofuranosyltransferase of C. glutamicum. Filled squares give positions in the Emb polypeptide where fusions resulted in alkaline phosphatase activity, with selected positions numbered (all fusions are given in the Supplementary data). The three numbered open squares give positions in the Emb polypeptide where fusions resulted in β-galactosidase activity. The numbers in the large rectangles give the amino acid positions delimiting the hydrophobic segments numbered I–XV. IN denotes the cytoplasmic side of the membrane and OUT the periplasmic side. Table I. Analysis of Cg-Emb topology using phoA-lacZ reporters Fusiona . Color . Phosphataseb, sp A (μmol·min−1·mg−1) . Galactosidaseb, sp A (μmol·min−1·mg−1) . Ratioc . Control No 0.30 0.08 – V67 Blue 3.75 0.31 12.1 W288 Blue 3.90 0.26 15.0 Y317 Blue 1.95 0.26 7.5 L395 Blue 1.72 0.70 2.5 G410 Blue 3.00 0.98 3.1 L454 Blue 1.46 0.24 6.1 L471 Red 0.76 1.90 0.4 D508 Blue 1.87 0.23 8.1 V565 Blue 1.93 0.27 7.1 R585 Red 0.75 3.21 0.2 M598 Blue 2.31 0.33 7.0 A615 Blue 2.20 0.41 5.4 A619 Blue 1.68 0.67 2.5 R637 Purple 0.89 3.34 0.3 W668 Blue 4.01 0.40 10.0 G677 Blue 2.50 0.25 10.0 Fusiona . Color . Phosphataseb, sp A (μmol·min−1·mg−1) . Galactosidaseb, sp A (μmol·min−1·mg−1) . Ratioc . Control No 0.30 0.08 – V67 Blue 3.75 0.31 12.1 W288 Blue 3.90 0.26 15.0 Y317 Blue 1.95 0.26 7.5 L395 Blue 1.72 0.70 2.5 G410 Blue 3.00 0.98 3.1 L454 Blue 1.46 0.24 6.1 L471 Red 0.76 1.90 0.4 D508 Blue 1.87 0.23 8.1 V565 Blue 1.93 0.27 7.1 R585 Red 0.75 3.21 0.2 M598 Blue 2.31 0.33 7.0 A615 Blue 2.20 0.41 5.4 A619 Blue 1.68 0.67 2.5 R637 Purple 0.89 3.34 0.3 W668 Blue 4.01 0.40 10.0 G677 Blue 2.50 0.25 10.0 Fusion points in Cg-Emb were determined by DNA sequencing and colony coloration on dual indicator plates judged after 48 h (Alexeyev and Winkler 1999). aPosition of the last residue of Cg-Emb followed by reporter. bEnzyme activities of the fusions, determined as described in Materials and methods, average of at least two independent experiments. cNormalized activity ratio rounded to the first decimal place. Open in new tab Table I. Analysis of Cg-Emb topology using phoA-lacZ reporters Fusiona . Color . Phosphataseb, sp A (μmol·min−1·mg−1) . Galactosidaseb, sp A (μmol·min−1·mg−1) . Ratioc . Control No 0.30 0.08 – V67 Blue 3.75 0.31 12.1 W288 Blue 3.90 0.26 15.0 Y317 Blue 1.95 0.26 7.5 L395 Blue 1.72 0.70 2.5 G410 Blue 3.00 0.98 3.1 L454 Blue 1.46 0.24 6.1 L471 Red 0.76 1.90 0.4 D508 Blue 1.87 0.23 8.1 V565 Blue 1.93 0.27 7.1 R585 Red 0.75 3.21 0.2 M598 Blue 2.31 0.33 7.0 A615 Blue 2.20 0.41 5.4 A619 Blue 1.68 0.67 2.5 R637 Purple 0.89 3.34 0.3 W668 Blue 4.01 0.40 10.0 G677 Blue 2.50 0.25 10.0 Fusiona . Color . Phosphataseb, sp A (μmol·min−1·mg−1) . Galactosidaseb, sp A (μmol·min−1·mg−1) . Ratioc . Control No 0.30 0.08 – V67 Blue 3.75 0.31 12.1 W288 Blue 3.90 0.26 15.0 Y317 Blue 1.95 0.26 7.5 L395 Blue 1.72 0.70 2.5 G410 Blue 3.00 0.98 3.1 L454 Blue 1.46 0.24 6.1 L471 Red 0.76 1.90 0.4 D508 Blue 1.87 0.23 8.1 V565 Blue 1.93 0.27 7.1 R585 Red 0.75 3.21 0.2 M598 Blue 2.31 0.33 7.0 A615 Blue 2.20 0.41 5.4 A619 Blue 1.68 0.67 2.5 R637 Purple 0.89 3.34 0.3 W668 Blue 4.01 0.40 10.0 G677 Blue 2.50 0.25 10.0 Fusion points in Cg-Emb were determined by DNA sequencing and colony coloration on dual indicator plates judged after 48 h (Alexeyev and Winkler 1999). aPosition of the last residue of Cg-Emb followed by reporter. bEnzyme activities of the fusions, determined as described in Materials and methods, average of at least two independent experiments. cNormalized activity ratio rounded to the first decimal place. Open in new tab Localization of fusions To locate the fusions in a topology model, a number of predictions were applied to Cg-Emb based on a variety of algorithms available at the Swiss Institute of Bioinformatics (data not shown). Method PHDhtm predicted the C-terminus to be cytosolic. This prediction was not considered due to the large number of active PhoA fusions between aa 756 and 1145, clearly indicating the C-terminus to be located towards the periplasmic side. Similarly, the TMHMM prediction was not considered, as was the case with other predictions due to their strong divergence from the experimental data. The highest consistency of the information derived from the fusions was obtained with the Dense Alignment Surface algorithm (Cserzö et al. 1997), which is based on low-stringency dot-plots of the query sequence against a collection of library sequences derived from nonhomologous membrane proteins. Almost all 62 fusions analyzed provide a best fit to the model shown in Figure 1. There are only a very limited number of exceptions, like the set of fusions at V719, R726, and E727. However, these fusions result in the absence of at least two of the Arg residues present in the original Emb protein in positions R726, R728, and R731. The absence of positively charged residues, dominating the cytosolic side of polytopic membrane proteins, is well known to disable proper membrane integration via the sec-translocon (Pohlschröder et al. 2005). Also, the fusion at position R637 replaces a positive Arg residue, which may thus lead in part to improper localization and explain the high phosphatase activity and vague colony color. In the case of phosphatase activity within transmembrane helical fusions, such as L454, V565, and A619, the mere length of the remaining hydrophobic stretch is expected to be too short to enable pulling of the reporter protein through the membrane to the cytosolic side. This is in accordance with a study on the E. coli Lac permease where at least half of a helix is necessary for insertion into the membrane (Calamia and Manoil 1990). Overall, there is a very good consistency of the experimental data with the model. Growth characteristics of Cg-emb mutants The availability of C. glutamicumΔemb, and its feature that solely due to plasmid-encoded Cg-emb the arabinan domain in mAGP is fully restored (Alderwick et al. 2005), is an ideal tool to study the consequences of Cg-emb mutations on Araf delivery to the mAGP complex. To this end, we introduced eight mutations into Cg-emb (Figure 2B) based on the mutations introduced into embC of M. smegmatis which affect LAM biosynthesis (Berg et al. 2005). These mutations together with their locations in the topological model are given in Table II. Cg-emb and alleles of this gene cloned in pEKEx2 were introduced into C. glutamicumΔemb and growth of the recombinant clones judged on plates (Table II). Whereas without emb present colonies of C. glutamicum Δemb were small with rough appearance, the single mutations W659L, P667A, I673A, and Q674E introduced into the Pro-motif of plasmid-encoded emb resulted in glossy large colonies characteristic for wild type Cg-emb. The triple mutation W627L–P667A–Q674E analyzed in the same manner produced colonies of intermediate size. D297A and D297A-D298A in the GT-C motif produced only tiny mate colonies indistinguishable from the negative control indicating a distortion of cell wall biosynthesis, whereas the D298A mutation resulted in colony size and surface characteristics of wild type Cg-emb. In addition, growth was analyzed in liquid culture. Again the positive control with plasmid-encoded emb in C. glutamicumΔemb exhibited growth comparable to the wild type, whereas growth rate and final optical clensity (OD) were significantly reduced with emb carrying the triple mutation (Figure 2A). The mutation D297A-D298A prevented growth, whereas D298A again restored growth (Figure 2A). Growth was also restored with the individual W659L, P667A, I673A, and Q674E mutations (data not shown). Interestingly, already truncation of the C-terminus by 159 amino acids of Cg-Emb resulted in inactive arabinosyltransferase as judged from the inability of C. glutamicumΔemb carrying the corresponding pEKEx2emb derivative to grow (data not shown). Fig. 2. Open in new tabDownload slide Mutations and growth analysis of mutant Emb arabinofuranosyltransferases in C. glutamicumΔemb. (A) Growth analysis due to mutated Emb arabinofuranosyltransferases in C. glutamicumΔemb containing the plasmid-encoded emb alleles. Growth was analyzed in BHI medium supplemented with 0.5 M sorbitol, followed by measuring the OD at 600 nm. ▪, controls of the wild type; ○, the mutant C. glutamicumΔemb without emb encoded arabinofuranosyltransferase activity (Alderwick et al. 2005); □, the same mutant with pEKEx2emb overexpressing emb. With the plasmid encoded emb carrying the D297A-D298A mutation (•) no complementation was obtained, whereas with the D298A mutation (▴), growth was almost identical as obtained with unmutated emb. With the triple mutation P659L–P667A–Q674E in emb (♦), an intermediate growth was obtained. (B) Sequence identities among the Emb proteins and mutations introduced in the periplasmic loop regions identified connecting helices III–IV, with the GT-C motif (Liu and Mushegian 2003) and XIII–XIV with the Pro-motif (Berg et al. 2005), respectively. The amino acid substitutions introduced in Cg-Emb are indicated by white letters on a black background. Beneath the alignments, the degree of conservation is indicated. Cglu stands for C. glutamicum, Mmar for M. marinum, Mtub for M. tuberculosis, and Mlep for M. leprae. Growth performance due the other mutations is given in Table II. Table II. Mutations in Cg-emb and consequences of colony formation of C. glutamicumΔemb complemented with Cg-emb alleles Strain . Mutation . Localization of mutation . Motif . Colony size of straina (mm) . C.g.Δemb pEKEx2 Negative control <0.4 C.g.Δemb pEKEx2emb Positive control 3 C.g.Δemb pEKEx2emb-D297A D297A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-D298A D298A Loops III–IV GT-C 3 C.g.Δemb pEKEx2emb-D297A-D298A D297A-D298A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-R350A R350A Helix IV 3 C.g.Δemb pEKEx2emb-K603A K603A Loops XI–XII 3 C.g.Δemb pEKEx2emb-W659L W659L Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-P667A P667A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-I673A I673A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-Q674E Q674E Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E W659L-P667A-Q674E Loops XIII–XIV Pro 1 C.g.Δemb pEKEx2emb-D1031A D1031A C-terminus 3 Strain . Mutation . Localization of mutation . Motif . Colony size of straina (mm) . C.g.Δemb pEKEx2 Negative control <0.4 C.g.Δemb pEKEx2emb Positive control 3 C.g.Δemb pEKEx2emb-D297A D297A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-D298A D298A Loops III–IV GT-C 3 C.g.Δemb pEKEx2emb-D297A-D298A D297A-D298A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-R350A R350A Helix IV 3 C.g.Δemb pEKEx2emb-K603A K603A Loops XI–XII 3 C.g.Δemb pEKEx2emb-W659L W659L Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-P667A P667A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-I673A I673A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-Q674E Q674E Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E W659L-P667A-Q674E Loops XIII–XIV Pro 1 C.g.Δemb pEKEx2emb-D1031A D1031A C-terminus 3 Most mutations are localized according to the model in periplasmic loop regions (Figure 1) and correspond to a GT-C and proline Pro-motif, respectively (Berg et al. 2005; Liu and Mushegian 2003). aPhenotype was determined after incubation for 1 week on plates containing BHI (Difco) plus 0.5 M sorbitol. Open in new tab Table II. Mutations in Cg-emb and consequences of colony formation of C. glutamicumΔemb complemented with Cg-emb alleles Strain . Mutation . Localization of mutation . Motif . Colony size of straina (mm) . C.g.Δemb pEKEx2 Negative control <0.4 C.g.Δemb pEKEx2emb Positive control 3 C.g.Δemb pEKEx2emb-D297A D297A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-D298A D298A Loops III–IV GT-C 3 C.g.Δemb pEKEx2emb-D297A-D298A D297A-D298A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-R350A R350A Helix IV 3 C.g.Δemb pEKEx2emb-K603A K603A Loops XI–XII 3 C.g.Δemb pEKEx2emb-W659L W659L Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-P667A P667A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-I673A I673A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-Q674E Q674E Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E W659L-P667A-Q674E Loops XIII–XIV Pro 1 C.g.Δemb pEKEx2emb-D1031A D1031A C-terminus 3 Strain . Mutation . Localization of mutation . Motif . Colony size of straina (mm) . C.g.Δemb pEKEx2 Negative control <0.4 C.g.Δemb pEKEx2emb Positive control 3 C.g.Δemb pEKEx2emb-D297A D297A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-D298A D298A Loops III–IV GT-C 3 C.g.Δemb pEKEx2emb-D297A-D298A D297A-D298A Loops III–IV GT-C <0.4 C.g.Δemb pEKEx2emb-R350A R350A Helix IV 3 C.g.Δemb pEKEx2emb-K603A K603A Loops XI–XII 3 C.g.Δemb pEKEx2emb-W659L W659L Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-P667A P667A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-I673A I673A Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-Q674E Q674E Loops XIII–XIV Pro 3 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E W659L-P667A-Q674E Loops XIII–XIV Pro 1 C.g.Δemb pEKEx2emb-D1031A D1031A C-terminus 3 Most mutations are localized according to the model in periplasmic loop regions (Figure 1) and correspond to a GT-C and proline Pro-motif, respectively (Berg et al. 2005; Liu and Mushegian 2003). aPhenotype was determined after incubation for 1 week on plates containing BHI (Difco) plus 0.5 M sorbitol. Open in new tab Analysis of cell wall associated lipids To obtain an initial phenotypic composition of the cell wall, the “free lipids” from each of the strains studied underwent lipid extraction and analysis by thin layer chromatography (TLC) (Figure 3). A clear increase in free trehalose monocorynomycolates (TMCM) could be observed in C. glutamicumΔemb, and the same strain carrying plasmid-encoded Cg-emb with the D297A mutation or the D297A-D298A mutation. A slight increase could be observed due to the P659L–P667A–Q674E mutation. This apparent increase in the amount of free TMCM in the cell wall is indicative of an altered cell wall ultrastructure, particularly when considering the removal of sites for mycolic acid attachment (McNeil et al. 1991; Alderwick et al. 2005; Alderwick, Dover et al. 2006). Although this view appears plausible, it could not be excluded that additional effects due to the, in part, altered growth characteristics of the mutants are present. Normal levels of TMCM could be observed with plasmids carrying the D298A, W659L, P667A, or the Q674E mutation, which is indicative of an intact cell wall and its associated free lipids. Fig. 3. Open in new tabDownload slide Analysis of cell wall associated lipids from C. glutamicum, C. glutamicumΔemb, and C. glutamicumΔemb carrying mutated emb alleles. Cell wall associated free lipids were analyzed as described in Materials and methods. The strains analyzed are as follows: 1, C. glutamicum; 2, C. glutamicumΔemb; 3, C. glutamicumΔemb pEKEx2emb-D297A; 4, C. glutamicumΔemb pEKEx2emb-D298A; 5, C. glutamicumΔemb pEKEx2emb-D297A-D298A; 6, C. glutamicumΔemb pEKEx2emb-W659L; 7, C. glutamicumΔemb pEKEx2emb-P667A; 8, C. glutamicumΔemb pEKEx2emb-Q674E; 9, C. glutamicumΔemb pEKEx2emb-W659L-P667A-Q674E. Cell wall bound corynomycolic acid compositional analysis It has previously been reported that a good indicator for the disruption of the arabinan component of the cell wall is the absence of AG-esterified corynomycolic acids (Alderwick et al. 2005; Alderwick, Dover et al. 2006). We therefore analyzed these bound lipids derivatized as corynomycolic acid methyl esters (CMAMEs) in C. glutamicumΔemb carrying the plasmid-encoded emb mutations. Corynebacterium glutamicum and C. glutamicum Δemb exhibited the expected presence and absence of CMAMEs, respectively (Figure 4, Lanes 1 and 2). No CMAMEs could be observed with the D297A mutation in Cg-emb (Figure 4, Lane 3). However, CMAMEs were detectable with the D298A mutation indicating the presence of mycolation sites (Figure 4, Lane 4), although the absolute amount of CMAMEs in comparison to C. glutamicum (Figure 4, Lane 1) hinted to a reduced number of possible CMAME-linkage sites. The D297A-D298A mutation resulted also in the absence of CMAMEs (Figure 4, Lane 5), thus confirming again that D297 is the more important amino acid residue within the GT-C motif. Although slightly reduced, the CMAMEs for the remaining mutants studied remained largely unaffected, apart from the triple mutant, which did exhibit a moderate decrease in the amount of apparent CMAMEs esterified in the cell wall (Figure 4, Lane 9). Fig. 4. Open in new tabDownload slide Analysis of cell wall bound CMAMEs of delipidated cells. An aliquot of the released cell wall bound CMAMEs from each strain was analyzed by TLC as described in Materials and methods. 1, C. glutamicum; 2, C. glutamicumΔemb; 3, C. glutamicumΔemb pEKEx2emb-D297A; 4, C. glutamicumΔemb pEKEx2emb-D298A; 5, C. glutamicumΔemb pEKEx2emb-D297A-D298A 2; 6, C. glutamicumΔemb pEKEx2emb-W659L; 7, C. glutamicumΔemb pEKEx2emb-P667A; 8, C. glutamicumΔemb pEKEx2emb-Q674E; 9, C. glutamicumΔemb pEKEx2emb-W659L-P667A-Q674E. Glycosyl compositional analysis To directly observe the effects of the mutated arabinosyltransferase Cg-emb on the arabinan domain structure, the mutants together with controls were subjected to glycosyl compositional analysis. Corynebacterium glutamicum and C. glutamicumΔemb afforded Ara:Gal ratios of 3.1 : 1 and 0.2 : 1, respectively (Figure 5, Lanes 1 and 2). The D297A mutation in emb produced a cell wall with a dramatically reduced arabinose content (Figure 5, Lane 3) almost identical to C. glutamicumΔemb. Also, with the D298A mutation, the level of arabinose was significantly reduced by approximately 80% to give an Ara:Gal ratio of 0.8 : 1 (Figure 5, Lane 4). The D297A-D298A mutation caused a phenotype identical to the D297A mutation, thus reinforcing the importance of the D297 residue in catalysis (Figure 5, Lane 5). Analysis of cell walls isolated from strains carrying mutations within the proline motif gave a varying profile of sugar composition (Figure 5, Lanes 6–9). W659L, P667A, and Q674E gave Ara:Gal ratios of 1.2:1, 1.5:1, and 1.9:1, respectively, whereas the triple mutation caused a further reduction of arabinose resulting in an Ara:Gal ratio of 0.56 : 1. In summary, there is a reduction of Ara in the mutants and therefore the Gal ratio relative to Ara is higher in the glycosyl compositional analysis. Moreover, it has to be borne in mind that the decreased Rha content with respect to Gal (Figure 5) can also be explained by the fact that we have additional Rha residues that are not in the linker region, i.e. not 4-linked Rha but are terminal Rha residues that are located at the terminal ends of the arabinan domains. So, the C. glutamicum emb deletion strain will retain a small amount of Rha as 4-Rha from the linker unit but will have decreased t-Rha. Fig. 5. Open in new tabDownload slide Glycosyl compositional analysis of cell walls from C. glutamicum, C. glutamicumΔemb, and C. glutamicumΔemb carrying mutated emb alleles. Samples were prepared and analyzed as described in Materials and methods. The data represent the relative sugar composition as a percentage in graphical format. The bars are tinted as follows: rhamnose, black; arabinose, white; galactose, hatched. 1, C. glutamicum; 2, C. glutamicumΔemb; 3, C. glutamicumΔemb pEKEx2emb-D297A; 4, C. glutamicumΔemb pEKEx2emb-D298A; 5, C. glutamicumΔemb pEKEx2emb-D297A-D298A; 6, C. glutamicumΔemb pEKEx2emb-W659L; 7, C. glutamicumΔemb pEKEx2emb-P667A; 8, C. glutamicumΔemb pEKEx2emb-Q674E; 9, C. glutamicumΔemb pEKEx2emb-W659L-P667A-Q674E. Glycosyl linkage analysis To further define the role of amino acid residues from Cg-Emb in the biosynthesis of cell wall arabinan, glycosidic linkages from cell walls isolated from C. glutamicumΔemb strains carrying mutations in plasmid-encoded Cg-emb were analyzed by GC/MS. For all such mutants studied, the 5-Galf, 6-Galf, and 5,6-Galf residues could be accounted for, which as expected represent an unmodified galactan domain (data not shown). As previously reported (Alderwick et al. 2005), deletion of Cg-emb resulted in a severely truncated arabinan domain with only single t-Araf residues attached to the galactan domain. The D298A mutation results in an increase in the relative t-Araf content with a reduced degree of branching while generally retaining other glycosidic linkages, which indicates abrogated Cg-Emb activity. However, the D297A and the D297A-D298A mutations fully abolished the Araf linkages of AG of C. glutamicum, excluding t-Araf (Table III). Thus, it is clear that D297 is a major catalytic residue in terms of arabinan deposition, albeit D298 also influences this activity. Typical glycosidic linkages in comparison with C. glutamicum are observed with each of the W627L, P667A, and Q674E mutations (Table III). Taken together with the reduced Ara:Gal ratio (Figure 5), this may imply that these residues in some way moderate the size of the arabinan chains in AG. This is substantiated by the fact that, when all three residues are mutated together, a severe reduction in arabinose can be observed by GC (Figure 5), but with all glycosidic linkages remaining present (Table III). Table III. Glycosyl linkage analysis of cell wall arabinan from C. glutamicum, C. glutamicum deleted of emb (C.g.Δemb), and the plasmid-encoded mutant alleles in the C.g.Δemb background as determined by GC/MS Strain . t-Araf . 2-Araf . 5-Araf . 3,5-Araf . 2,5-Araf . C. glutamicum 8.7 9.3 52.9 11.9 17.2 C.g.Δemb 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D297A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D298A 20.6 9.9 45.4 13.4 10.7 C.g.Δemb pEKEx2emb-D297A-D298A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-W659L 16.7 10.8 51.7 10.3 10.3 C.g.Δemb pEKEx2emb-P667A 13.7 11.4 51.0 11.2 12.7 C.g.Δemb pEKEx2emb-Q674E 13.9 6.1 55.1 12.8 12.1 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E 34.2 9.3 45.8 6.2 4.5 Strain . t-Araf . 2-Araf . 5-Araf . 3,5-Araf . 2,5-Araf . C. glutamicum 8.7 9.3 52.9 11.9 17.2 C.g.Δemb 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D297A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D298A 20.6 9.9 45.4 13.4 10.7 C.g.Δemb pEKEx2emb-D297A-D298A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-W659L 16.7 10.8 51.7 10.3 10.3 C.g.Δemb pEKEx2emb-P667A 13.7 11.4 51.0 11.2 12.7 C.g.Δemb pEKEx2emb-Q674E 13.9 6.1 55.1 12.8 12.1 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E 34.2 9.3 45.8 6.2 4.5 Values are given as relative glycosyl linkages in percentage. Open in new tab Table III. Glycosyl linkage analysis of cell wall arabinan from C. glutamicum, C. glutamicum deleted of emb (C.g.Δemb), and the plasmid-encoded mutant alleles in the C.g.Δemb background as determined by GC/MS Strain . t-Araf . 2-Araf . 5-Araf . 3,5-Araf . 2,5-Araf . C. glutamicum 8.7 9.3 52.9 11.9 17.2 C.g.Δemb 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D297A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D298A 20.6 9.9 45.4 13.4 10.7 C.g.Δemb pEKEx2emb-D297A-D298A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-W659L 16.7 10.8 51.7 10.3 10.3 C.g.Δemb pEKEx2emb-P667A 13.7 11.4 51.0 11.2 12.7 C.g.Δemb pEKEx2emb-Q674E 13.9 6.1 55.1 12.8 12.1 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E 34.2 9.3 45.8 6.2 4.5 Strain . t-Araf . 2-Araf . 5-Araf . 3,5-Araf . 2,5-Araf . C. glutamicum 8.7 9.3 52.9 11.9 17.2 C.g.Δemb 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D297A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-D298A 20.6 9.9 45.4 13.4 10.7 C.g.Δemb pEKEx2emb-D297A-D298A 100 0.0 0.0 0.0 0.0 C.g.Δemb pEKEx2emb-W659L 16.7 10.8 51.7 10.3 10.3 C.g.Δemb pEKEx2emb-P667A 13.7 11.4 51.0 11.2 12.7 C.g.Δemb pEKEx2emb-Q674E 13.9 6.1 55.1 12.8 12.1 C.g.Δemb pEKEx2emb-W659L-P667A-Q674E 34.2 9.3 45.8 6.2 4.5 Values are given as relative glycosyl linkages in percentage. Open in new tab Discussion The mAGP complex represents one of the most important cell wall components of the Corynebacterianeae and is essential for the viability of M. tuberculosis (Vilcheze et al. 2000; Pan et al. 2001; Gande et al. 2004; Mills et al. 2004). It is therefore not surprising that one of the most effective antimycobacterial drugs, EMB, targets its biosynthesis. An elementary structure of this mAGP polymer is apparent in C. glutamicum, and this bacterium has proved useful for a number of studies on mAGP biosynthesis (Gande et al. 2004; Portevin et al. 2004; Alderwick et al. 2005; Alderwick, Seidel et al. 2006). It represents the archetype of the Corynebacterianeae with respect to a core cell wall structure and also a low number of gene duplications (Nakamura et al. 2003) when compared with M. tuberculosis which has a much more elaborated cell wall and also a significantly larger genome. Although, arabinofuranosyltransferases are of prime interest for arabinan formation in Corynebacterianeae, functional and structural studies of these glycosyltransferases are scarce. The sequences of Mt-EmbC, Mt-EmbA, Mt-EmbB, and Cg-Emb exhibit a high degree of similarity among each other (54–58%) (Berg et al. 2005), with Mt-EmbC probably closest to the single Cg-Emb. There are two domains predicted for the Emb proteins. They consist of a hydrophilic C-terminus and a hydrophobic N-terminus as typical for membrane-embedded proteins. Predictions of the hydrophobic part of Mt-EmbC range from 11 to 14 hydrophobic segments (HSs) which may span the membrane (Berg et al. 2005). Interestingly, in all the predictions we compared (data not shown), these were largely on par with respect to HSs corresponding to helices I–III and VII-XV of our current model (Figure 1), whereas for the regions in between variations existed. The fusions obtained (Table 1 and Supplementary Table S1) are consistent with three HSs in this part, because the large number of fusions obtained with PhoA activity of the entire remaining Emb protein would have to be located on the cytoplasmic side. In addition, a single HS does not match the experimental data due to active PhoA fusions at L395, P396, and G410. This allows a clear assignment of specific loop regions and the C-terminus to the periplasmic side as shown in Figure 1. As HS I has the strongest hydrophobicity score (Cuff et al. 1998) of all the HSs, we regard this as membrane spanning and not as a segment located in the periplasm. Application of a segment-based multiple sequence alignment (Subramanian et al. 2005) identified particularly conserved regions within the Emb proteins (data not shown). The highest conservation is present in the segment covering aa 286–410 of Mt-EmbC, which is consistent in part with the GT-C motif of glycosyltransferases (Liu and Mushegian 2003). The clear assignment of HSs and their orientation enables a further delineation of the structure–function aspects of the Emb proteins. Within HSs IV, the fully conserved motif WMRLP (343–347 aa in Mt-EmbC) is present. It is known that proline residues in membrane spanning helices introduce kink angles of about 20° which may position functionally important residues in the three-dimensional structure (Grigorieff et al. 1996). For example, Pro50 in helix B in bacteriorhodopsin positions an unpaired carbonyl oxygen of Thr46 which forms part of the channel (Deisenhofer and Michel 1989). Therefore, it is very likely that this part of Emb is of functional importance. A further indication is the fact that the adjacent HSs V and VI are short or represent loop regions that dip into the membrane. As the three-dimensional structures of membrane proteins, such as that of either aquaporin or the glutamate transporter, illustrate (Yernool et al. 2004), there are regions which do not simply form transmembrane spanning HSs but which span only a portion of the membrane which are mechanistically of great significance. Thus, the charged residues in the motif NGLRPE between HSs V and VI (394–399 aa in Mt-EmbC) could be involved in translocation of a substrate related to EMB function. We recently identified AftA as a new mycobacterial arabinosyltransferase, which like the Emb proteins incorporates single Araf residue onto the galactan domain (Alderwick, Seidel et al. 2006). The introduction of these Araf residues by Emb and AftA activity is fully dependent on the donor decaprenol phosphoarabinose (DPA). Although some similarities between AftA and the Emb proteins exist (Alderwick, Seidel et al. 2006), our sequence and topology comparison did not reveal a region obvious for translocation of the assumed DPA precursor but could, for instance, be related to the accepting glycosyl unit(s). In terms of catalytic mechanism, the details of arabinofuranosyl transfer to the accepting polymer remain to be defined. There is a lack of experimental data regarding the mechanism of glycosyl transfer within the GT-C super family of glycosyltransferases, probably due to the fact that these are large membrane proteins and that their function within Corynebacterianeae is only now becoming clearer. A conserved element of these proteins is a modified DXD motif (Liu and Mushegian 2003), which is DDG in the Emb proteins (Figure 2B), which we are now able to locate between the third and fourth HS. Within this motif, the first Asp residue, D297, is clearly the most important residue for a fully functional Emb protein, with the second Asp, D298, having a less pronounced effect on arabinose deposition and linkage establishment of AG. In a study with a eukaryotic mannosyltransferase of Trypanosoma brucei, both Asp residues of the DXD motif have been demonstrated to be of comparable importance for glycosyltransferase activity (Maeda et al. 2001). As both Asp residues are adjacent in the Emb proteins, one might expect the functional group of these residues to face in opposite directions; therefore, the notion of these acidic residues being involved in sugar–phosphate coordination via a divalent metal ion requires further clarification (Liu and Mushegian 2003). Mutagenic analysis of the amino acids, in the proposed proline motif of Cg-emb (Figure 2B) (Berg et al. 2005), resulted in indistinct phenotypes. The mutation of W659L caused an overall decrease in CMAMEs and arabinose but resulted in a cell wall retaining all AG-specific glycosyl linkages. The same phenotype could be observed for P667L and Q674E, but with differing levels of glycosidic linkages, but again with the retention of a wild type glycosidic linkage profile. The triple W659L–P667A–Q674E mutation, however, did result in a marked reduction in arabinose and CMAMEs, with a concomitant increase in cell wall associated TMCM. This suggests that the residues of this periplasmic loop region are clearly implicated in the coordination of polysaccharide chain length formation or regulation, respectively, rather than the maturation of the terminal Ara6 motif (Morona et al. 1995; Becker and Pühler 1998; Daniels and Morona 1999; Oriol et al. 2002; Berg et al. 2005). The fact that the glycosyl linkage types of all mutants are largely retained could indicate that Cg-Emb and possibly also the other Emb proteins are mostly involved in establishing a linear Ara chain as that in mature AG attached to the single Ara residue introduced by AftA. It has not gone unnoticed that the occurrence of several other arabinofuranosyltransferases could exist which work in concert with the Emb protein in the biosynthesis of AG. Clearly, the mode of arabinan decoration requires further investigation, because it cannot be expected that a single arabinofuranosyltransferase (Cg-Emb) would be responsible for such a plethora of arabinofuranosyl glycosidic linkages, including the C. glutamicum specific 2,5-Araf linkages. The identification of new cell wall biosynthetic drug targets is of great importance, especially with the emergence of multidrug-resistant tuberculosis. As a result, a clearer understanding of the Emb-related DPA-dependent arabinofuranosyltransferase structure and function is of paramount importance for the further exploitation of potential drug targets to disrupt the essential mAGP complex in mycobacterial species such as M. tuberculosis. Materials and methods Strains and culture conditions Corynebacterium glutamicum ATCC 13032 (the wild type strain) and C. glutamicumΔemb (Alderwick et al. 2005) together with its recombinant derivatives with emb alleles were grown on brain–heart infusion (BHI, Difco, Detroit, MI) containing 0.5 M sorbitol at 30°C (Eggeling and Bott 2005). Escherichia coli DH5α was grown on Luria-Bertani (LB) medium (Difco) at 37°C. When appropriate, kanamycin was used at a concentration of 25 µg/mL and ampicillin at a concentration of 50 µg/mL. Samples for cell wall analyses of C. glutamicum strains were prepared by harvesting cells at an OD of 5–8, followed by a saline wash and freeze-drying. Construction of emb-phoA–lacZ fusion and exonuclease treatment To make vector pMS3 suitable for exonuclease treatment of Cg-emb, pMA632 (Alexeyev and Winkler 1999) was NruI/NraI digested and ligated with a mixture of CGACCGCGGGG (all oligonucleotides are given in direction 5′ to 3′) and GGCGCCCCGCGGTCG to generate the new restriction site KspI in vector pMA632-1. This vector was prepared by a SmaI–KspI digest and treated with alkaline phosphatase. To obtain emb with appropriate restriction sites attached, pEKEx2emb was ScaI/EcoRI cleaved and ligated with a mixture of oligonucleotide CGACCGCGGGG and GGCGCCCCGCGGTCG to generate pEKEx2emb-2. From this vector, emb was obtained as a SwaI–KspI fragment that was purified and ligated with pMA632-1 to generate pMS3. The integrity of the emb-phoA–lacZ fusion was verified by sequencing. To construct progressive unidirectional deletions of the 3′ end of emb, pMS3 was cleaved with FseI–ScaI and treated for various times with exonuclease III (Henikoff 1984). Subsequently, single-stranded protruding DNA ends were digested with nuclease S1 and DNA blunted and religated. The ligation mixture was electroporated into E. coli DH5α and plated on dual indicator plates containing blue alkaline phosphatase and red β-galactosidase chromogenic substrates (Alexeyev and Winkler 1999). The plates contained 1.5% Bacto-agar, 1% Bacto-tryptone, 0.5% yeast extract, 1% NaCl, 80 µg/mL 5-bromo-4-chloro-3-indolyl-β-d-galactoside disodium salt (Roche, Palo Alto, CA), 100 µg/mL, chloro-3-indolyl-β-d-galactopyranoside (rose-Gal, AppliChem Inc., Cheshire, CT), and 50 µg/mL kanamycin. The plasmid DNA was isolated from different colored colonies and insertion points mapped by restriction digests and confirmed by sequencing. Site-directed mutagenesis Using pEKEx2emb as a template (Radmacher et al. 2005), mutations in emb were introduced in a two-step polymerase chain reaction (PCR) (Landt et al. 1990). The following mutation primers were used (with the mutated codon underlined): For D297A, AACACCTCTGCCGACGGCTTCATC; D298A, AACACCTCTGACGCCGGCTTCATC, D297A-D298A, AACACCTCTGCCGCCGGCTTCATC, in all cases with CGCAAGTAGAGCTCCCATCGC as the identical second primer. The resulting megaprimers were purified and used with GCTTGCATGCCTGCAGGTCGA in a second PCR to yield mutated emb fragments. These were treated with SbfI and SacI and ligated with pEKEx2emb from which the GT-C site containing SbfI–SacI fragment had been previously removed. Similarly, mutations were introduced into the AspI–ScaI fragment of emb using primers: W659L, GGAGGTGTACAAGAATCCGTT; P667A, ATCCCACCATGCCACGGCGTA; I673A, CTTGATCTGGGCGGTTTTATC; Q674E, CTTGATCTCGATGGTTTTATCCC. The second primer in the first PCR was CGATCAGACTCTGTCAACCGT, and the additional primer for the second PCR was TCCGGTTCCAGTACTGAAGGT. After cloning, the integrity of all replaced fragments with their adjacent sites was verified by sequencing. The triple mutation was obtained commercially from 4base lab, Reutlingen, Germany. Enzyme assays For alkaline phosphatase and β-galactosidase activity determinations, 1 mL LB overnight cultures of E. coli cells bearing fusion constructs and unfused plasmid (background control) were harvested by centrifugation, washed, and resuspended in 10 mM Tris–HCl, 10 mM MgSO4, pH 8 to an OD 600 of 1. Cells in 100 µL of this suspension were permeabilized by addition of 50 µL 0.1% sodium dodecyl sulfate (SDS) and 50 µL of chloroform. After incubation for 5 min at 37 °C, cells were stored on ice for enzyme activity determinations. Alkaline phosphatase activity was assayed essentially as described by Brickman and Beckwith (1975), except using an extinction coefficient of ε405 nm = 1.85 × 104 M−1cm−1 with 5-bromo-4-chloro-3-indolyl phosphate disodium salt as a substrate, whereas β-galactosidase was assayed according to Miller (1992) with nitrophenyl-β-galactoside using an extinction coefficient of ε420 nm = 2.13 × 104 M−1 cm−1. Protein was determined by the bicinchoninic method, and specific activities expressed in micromoles per minute per milligram. Extraction and analysis of cell wall associated and cell wall bound lipids Cells were grown as described earlier, harvested, washed, and freeze-dried. Cells (100 mg) were extracted by two consecutive extractions using 2 ml of CHCl3/CH3OH/H2O (10:10:3, v/v/v) for 3 h at 50 °C and the resulting delipidated cells stored for further use (as described subsequently). Organic extracts were combined with 1.75 mL CHCl3 and 0.75 mL H2O, mixed and centrifuged. The lower organic phase was recovered, washed twice with 2 mL of CHCl3/CH3OH/H2O (3:47:48, v/v/v), dried, and resuspended in 200 µL of CHCl3/CH3OH/H2O (10:10:3, v/v/v). An aliquot (20 µL) was analyzed by TLC using silica gel plates (5735 silica gel 60F254, Merck, Darmstadt, Germany) developed in CHCl3/CH3OH/H2O (60:16:2, v/v/v). TLCs were visualized by charring with 5% molybdophosphoric acid in ethanol at 100°C to reveal cell wall associated lipids. The bound lipids from delipidated extracts or purified cell walls (Isolation of the mAGP complex) were released by the addition of a 5% aqueous solution of tetra-butyl ammonium hydroxide, followed by overnight incubation at 100°C, and methylated as described in Alderwick et al. (2005). CMAMEs were analyzed by TLC using silica gel plates (5735 silica gel 60F254, Merck) developed in petroleum ether/acetone (95:5, v/v). TLCs were visualized by charring with 5% molybdophosphoric acid in ethanol at 100°C to reveal CMAMEs. Isolation of the mAGP complex The thawed cells were resuspended in phosphate-buffered saline containing 2% Triton X-100 (pH 7.2), disrupted by sonication, and centrifuged at 27 000g (Daffe et al. 1990; Besra et al. 1995). The pelleted material was extracted three times with 2% SDS in phosphate-buffered saline at 95°C for 1 h to remove associated proteins, successively washed with water, 80% (v/v) acetone in water, and acetone, and finally lyophilized to yield a highly purified cell wall preparation (Besra et al. 1995; Alderwick et al. 2005). Glycosyl composition and linkage analysis of cell walls Cell wall preparations were hydrolyzed using 2 M trifluoroacetic acid (TFA), reduced with NaB2H4 and the resultant alditols per-O-acetylated and examined by GC as described in Daffe et al. (1990), Besra et al. (1995), and Alderwick et al. (2005). Cell wall preparations were per-O-methylated using dimethyl sulfinyl carbanion (Daffe et al. 1990; Besra et al. 1995; Alderwick et al. 2005). The per-O-methylated cell walls were hydrolyzed using 2 M TFA, reduced with NaB2H4, per-O-acetylated, and examined by GC/MS. Analysis of alditol acetate sugar derivatives was performed on a CE Instruments ThermoQuest Trace GC 2000 in the splitless mode using a DB225 column (Supelco, Pennsylvania, PA). The oven was programmed to hold at an isothermal temperature of 275 °C for a run time of 15 min (Alderwick et al. 2005). GC/MS was carried out on a Finnigan Polaris/GCQ PlusTM using a BPX5 column (Supelco). Acknowledgments L.J.A. and M.S. contributed equally to this work. L.J.A. is a Biotechnology and Biological Sciences Research Council Quota Student. G.S.B. acknowledges support in the form of a Personal Research chair from Mr James Bardrick and as a former Lister Institute-Jenner Research Fellow, the Medical Research Council (UK). H.S. acknowledges the support from the Fonds der Chemischen Industrie. We thank Graham Burns and Karin Krumbach for technical assistance. Conflict of interest statement None declared. Abbreviations AG arabinogalactan Ara arabinose BHI brain–heart infusion CMAME corynomycolic acid methyl ester DPA decaprenol phosphoarabinose EMB ethambutol Gal galactose GC gas chromatography GT-C glycosyltransferase-C HS hydrophobic segment LAM lipoarabinomannan LB Luria-Bertani mAGP mycolyl-arabinogalactan–peptidoglycan MS mass spectrometry OD optical density PCR polymerase chain reaction Rha rhamnose SDS sodium dodecyl sulfate TFA trifluoroacetic acid TLC thin layer chromatography TMCM trehalose monocorynomycolates References Alderwick LJ , Dover L , Seidel M , Gande R , Sahm H , Eggeling L , Besra GS . Arabinan deficient mutants of Corynebacterium glutamicum and the consequent flux in decaprenylmonophosphoryl-d-arabinose metabolism , Glycobiology , 2006 , vol. 16 (pg. 1073 - 1081 ) Google Scholar Crossref Search ADS PubMed WorldCat Alderwick LJ , Radmacher E , Seidel M , Gande R , Hitchen PG , Morris HR , Dell A , Sahm H , Eggeling L , Besra GS . Deletion of Cg-emb in Corynebacterianeae leads to a novel truncated cell wall arabinogalactan, whereas inactivation of Cg-ubiA results in an arabinan-deficient mutant with a cell wall galactan core , J Biol Chem , 2005 , vol. 280 (pg. 32362 - 32371 ) Google Scholar Crossref Search ADS PubMed WorldCat Alderwick LJ , Seidel M , Sahm H , Besra GS , Eggeling L . Identification of a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan biosynthesis in Mycobacterium tuberculosis , J Biol Chem , 2006 , vol. 281 (pg. 15653 - 15661 ) Google Scholar Crossref Search ADS PubMed WorldCat Alexeyev MF , Winkler HH . Membrane topology of the Rickettsia prowazekii ATP/ADP translocase revealed by novel dual pho-lac reporters , J Mol Biol , 1999 , vol. 285 (pg. 1503 - 1513 ) Google Scholar Crossref Search ADS PubMed WorldCat Becker A , Pühler A . Specific amino acid substitutions in the proline-rich motif of the Rhizobium meliloti ExoP protein result in enhanced production of low-molecular-weight succinoglycan at the expense of high-molecular-weight succinoglycan , J Bacteriol , 1998 , vol. 180 (pg. 395 - 399 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat Berg S , Starbuck J , Torrelles JB , Vissa V , Crick DC , Chatterjee D , Brennan PJ . Roles of conserved proline and glycosyltransferase motifs of EmbC in biosynthesis of lipoarabinomannan , J Biol Chem , 2005 , vol. 280 (pg. 5651 - 5663 ) Google Scholar Crossref Search ADS PubMed WorldCat Besra GS , Khoo KH , McNeil MR , Dell A , Morris HR , Brennan PJ . A new interpretation of the structure of the mycolyl-arabinogalactan complex of Mycobacterium tuberculosis as revealed through characterization of oligoglycosylalditol fragments by fast-atom bombardment mass spectrometry and 1H nuclear magnetic resonance spectroscopy , Biochemistry , 1995 , vol. 34 (pg. 4257 - 4266 ) Google Scholar Crossref Search ADS PubMed WorldCat Bloom BR , Murray CJ . Tuberculosis: commentary on a reemergent killer , Science , 1992 , vol. 257 (pg. 1055 - 1064 ) Google Scholar Crossref Search ADS PubMed WorldCat Brickman E , Beckwith J . Analysis of the regulation of Escherichia coli alkaline phosphatase synthesis using deletions and phi80 transducing phages , J Mol Biol , 1975 , vol. 96 (pg. 307 - 316 ) Google Scholar Crossref Search ADS PubMed WorldCat Calamia J , Manoil C . Lac permease of Escherichia coli: topology and sequence elements promoting membrane insertion , Proc Natl Acad Sci USA , 1990 , vol. 87 (pg. 4937 - 4941 ) Google Scholar Crossref Search ADS PubMed WorldCat Chatterjee D , Hunter SW , McNeil M , Brennan PJ . Lipoarabinomannan. Multiglycosylated form of the mycobacterial mannosylphosphatidylinositols , J Biol Chem , 1992 , vol. 267 (pg. 6228 - 33 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat Cserzö M , Wallin E , Simon I , von Heijne G , Elofsson A . Prediction of transmembrane alpha-helices in prokaryotic membrane proteins: the dense alignment surface method , Protein Eng , 1997 , vol. 10 (pg. 673 - 676 ) Google Scholar Crossref Search ADS PubMed WorldCat Cuff JA , Clamp ME , Siddiqui AS , Finlay M , Barton GJ . JPred: a consensus secondary structure prediction server , Bioinformatics , 1998 , vol. 14 (pg. 892 - 893 ) Google Scholar Crossref Search ADS PubMed WorldCat Daffe M , Brennan PJ , McNeil M . Predominant structural features of the cell wall arabinogalactan of Mycobacterium tuberculosis as revealed through characterization of oligoglycosyl alditol fragments by gas chromatography/mass spectrometry and by 1H and 13C NMR analyses , J Biol Chem , 1990 , vol. 265 (pg. 6734 - 6743 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat Daniels C , Morona R . Analysis of Shigella flexneri wzz (Rol) function by mutagenesis and cross-linking: wzz is able to oligomerize , Mol Microbiol , 1999 , vol. 34 (pg. 181 - 194 ) Google Scholar Crossref Search ADS PubMed WorldCat Deisenhofer J , Michel H . Nobel lecture. The photosynthetic reaction centre from the purple bacterium Rhodopseudomonas viridis , EMBO J , 1989 , vol. 8 (pg. 2149 - 2170 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat Eggeling L , Bott M . Handbook of Corynebacterium glutamicum. , 2005 Boca Raton (FL) CRC Press, Taylor Francis Group Google Scholar Escuyer VE , Lety MA , Torrelles JB , Khoo KH , Tang J , Rithner CD , Frehel C , McNei MR , Brennan PJ , Chatterjee D . The role of the embA and embB gene products in the biosynthesis of the terminal hexaarabinofuranosyl motif of Mycobacterium smegmatis arabinogalactan , J Biol Chem , 2001 , vol. 276 (pg. 48854 - 48862 ) Google Scholar Crossref Search ADS PubMed WorldCat Gande R , Gibson KJ , Brown AK , Krumbach K , Dover LG , Sahm H , Shioyama S , Oikawa T , Besra GS , Eggeling L . Acyl-CoA carboxylases (accD2 and accD3), together with a unique polyketide synthase (Cg-pks), are key to mycolic acid biosynthesis in Corynebacterianeae such as Corynebacterium glutamicum and Mycobacterium tuberculosis , J Biol Chem , 2004 , vol. 279 (pg. 44847 - 44857 ) Google Scholar Crossref Search ADS PubMed WorldCat Gibson KJ , Eggeling L , Maughan WN , Krumbach K , Gurcha SS , Nigou J , Puzo G , Sahm H , Besra GS . Disruption of Cg-Ppm1, a polyprenyl monophosphomannose synthase, and the generation of lipoglycan-less mutants in Corynebacterium glutamicum. , J Biol Chem , 2003 , vol. 278 (pg. 40842 - 40850 ) Google Scholar Crossref Search ADS PubMed WorldCat Grigorieff N , Ceska TA , Downing KH , Baldwin JM , Henderson R . Electron-crystallographic refinement of the structure of bacteriorhodopsin , J Mol Biol , 1996 , vol. 259 (pg. 393 - 421 ) Google Scholar Crossref Search ADS PubMed WorldCat Henikoff S . Unidirectional digestion with exonuclease III creates targeted breakpoints for DNA sequencing , Gene , 1984 , vol. 28 (pg. 351 - 359 ) Google Scholar Crossref Search ADS PubMed WorldCat Kalinowski J , Bathe B , Bartels D , Bischoff N , Bott M , Burkovski A , Dusch N , Eggeling L , Eikmanns BJ , Gaigalat L , et al. The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of l-aspartate-derived amino acids and vitamins , J Biotechnol , 2003 , vol. 104 (pg. 5 - 25 ) Google Scholar Crossref Search ADS PubMed WorldCat Kremer L , Dover LG , Morehouse C , Hitchin P , Everett M , Morris HR , Dell A , Brennan PJ , McNeil MR , Flaherty C , et al. Galactan biosynthesis in Mycobacterium tuberculosis. Identification of a bifunctional UDP-galactofuranosyltransferase , J Biol Chem , 2001 , vol. 276 (pg. 26430 - 26440 ) Google Scholar Crossref Search ADS PubMed WorldCat Landt O , Grunert HP , Hahn U . A general method for rapid site-directed mutagenesis using the polymerase chain reaction , Gene , 1990 , vol. 96 (pg. 125 - 128 ) Google Scholar Crossref Search ADS PubMed WorldCat Liu J , Mushegian A . Three monophyletic superfamilies account for the majority of the known glycosyltransferases , Protein Sci , 2003 , vol. 12 (pg. 1418 - 1431 ) Google Scholar Crossref Search ADS PubMed WorldCat Maeda Y , Watanabe R , Harris CL , Hong Y , Ohishi K , Kinoshita K , Kinoshita T . PIG-M transfers the first mannose to glycosylphosphatidylinositol on the lumenal side of the ER , EMBO J , 2001 , vol. 20 (pg. 250 - 261 ) Google Scholar Crossref Search ADS PubMed WorldCat McNeil M , Daffe M , Brennan PJ . Evidence for the nature of the link between the arabinogalactan and peptidoglycan of mycobacterial cell walls , J Biol Chem , 1990 , vol. 265 (pg. 18200 - 18206 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat McNeil M , Daffe M , Brennan P . Location of the mycolyl ester substituents in the cell walls of mycobacteria , J Biol Chem , 1991 , vol. 266 (pg. 13217 - 13223 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat Mikusova K , Yagi T , Stern R , McNeil MR , Besra GS , Crick DC , Brennan PJ . Biosynthesis of the galactan component of the mycobacterial cell wall , J Biol Chem , 2000 , vol. 275 (pg. 33890 - 33897 ) Google Scholar Crossref Search ADS PubMed WorldCat Miller JH . A Short Course in Bacterial Genetics , 1992 Cold Spring Harbor (NY) Cold Spring Harbor Laboratory Google Scholar Mills JA , Motichka K , Jucker M , Wu HP , Uhlik BC , Stern RJ , Scherman MS , Vissa VD , Pan F , Kundu M , et al. Inactivation of the mycobacterial rhamnosyltransferase, which is needed for the formation of the arabinogalactan-peptidoglycan linker, leads to irreversible loss of viability , J Biol Chem , 2004 , vol. 279 (pg. 43540 - 43546 ) Google Scholar Crossref Search ADS PubMed WorldCat Morona R , van den Bosch L , Manning PA . Molecular, genetic, and topological characterization of O-antigen chain length regulation in Shigella flexneri , J Bacteriol , 1995 , vol. 177 (pg. 1059 - 1068 ) Google Scholar Crossref Search ADS PubMed WorldCat Nakamura Y , Nishio Y , Ikeo K , Gojobori T . The genome stability in Corynebacterium species due to lack of the recombinational repair system , Gene , 2003 , vol. 317 (pg. 149 - 155 ) Google Scholar Crossref Search ADS PubMed WorldCat Oriol R , Martinez-Duncker I , Chantret I , Mollicone R , Codogno P . Common origin and evolution of glycosyltransferases using Dol-P-monosaccharides as donor substrate , Mol Biol Evol , 2002 , vol. 19 (pg. 1451 - 1463 ) Google Scholar Crossref Search ADS PubMed WorldCat Pan F , Jackson M , Ma Y , McNeil M . Cell wall core galactofuran synthesis is essential for growth of mycobacteria , J Bacteriol , 2001 , vol. 183 (pg. 3991 - 3998 ) Google Scholar Crossref Search ADS PubMed WorldCat Pohlschröder M , Hartmann E , Hand NJ , Dilks K , Haddad A . Diversity and evolution of protein translocation , Annu Rev Microbiol , 2005 , vol. 59 (pg. 91 - 111 ) Google Scholar Crossref Search ADS PubMed WorldCat Portevin D , De Sousa-D'Auria C , Houssin C , Grimaldi C , Chami M , Daffe M , Guilhot C . A polyketide synthase catalyzes the last condensation step of mycolic acid biosynthesis in mycobacteria and related organisms , Proc Natl Acad Sci USA , 2004 , vol. 101 (pg. 314 - 319 ) Google Scholar Crossref Search ADS PubMed WorldCat Radmacher E , Stansen KC , Besra GS , Alderwick LJ , Maughan WN , Hollweg G , Sahm H , Wendisch VF , Eggeling L . Ethambutol, a cell wall inhibitor of Mycobacterium tuberculosis, elicits l-glutamate efflux of Corynebacterium glutamicum , Microbiology , 2005 , vol. 151 (pg. 1359 - 1368 ) Google Scholar Crossref Search ADS PubMed WorldCat Ramaswamy SV , Amin AG , Goksel S , Stager CE , Dou SJ , El Sahly H , Moghazeh SL , Kreiswirth BN , Musser JM . Molecular genetic analysis of nucleotide polymorphisms associated with ethambutol resistance in human isolates of Mycobacterium tuberculosis , Antimicrob Agents Chemother , 2000 , vol. 44 (pg. 326 - 336 ) Google Scholar Crossref Search ADS PubMed WorldCat Shi L , Berg S , Lee A , Spencer JS , Zhang J , Vissa V , McNeil MR , Khoo KH , Chatterjee D . The carboxy terminus of EmbC from Mycobacterium smegmatis mediates chain length extension of the arabinan in lipoarabinomannan , J Biol Chem , 2006 , vol. 281 (pg. 19512 - 19526 ) Google Scholar Crossref Search ADS PubMed WorldCat Subramanian AR , Weyer-Menkhoff J , Kaufmann M , Morgenstern B . DIALIGN-T: an improved algorithm for segment-based multiple sequence alignment BMC , Bioinformatics , 2005 , vol. 6 pg. 66 Google Scholar PubMed OpenURL Placeholder Text WorldCat Takayama K , Kilburn JO . Inhibition of synthesis of arabinogalactan by ethambutol in Mycobacterium smegmatis , Antimicrob Agents Chemother , 1989 , vol. 33 (pg. 1493 - 1499 ) Google Scholar Crossref Search ADS PubMed WorldCat Telenti A , Philipp WJ , Sreevatsan S , Bernasconi C , Stockbauer KE , Wieles B , Musser JM , Jacobs WR Jr . The emb operon, a gene cluster of Mycobacterium tuberculosis involved in resistance to ethambutol , Nat Med , 1997 , vol. 3 (pg. 567 - 570 ) Google Scholar Crossref Search ADS PubMed WorldCat Vilcheze C , Morbidoni HR , Weisbrod TR , Iwamoto H Kuo M , Sacchettini JC , Jacobs WR Jr . Inactivation of the inhA-encoded fatty acid synthase II (FASII) enoyl-acyl carrier protein reductase induces accumulation of the FASI end products and cell lysis of Mycobacterium smegmatis , J Bacteriol , 2000 , vol. 182 (pg. 4059 - 4067 ) Google Scholar Crossref Search ADS PubMed WorldCat Yernool D , Boudker O , Jin Y , Gouaux E . Structure of a glutamate transporter homologue from Pyrococcus horikoshii , Nature , 2004 , vol. 431 (pg. 811 - 818 ) Google Scholar Crossref Search ADS PubMed WorldCat Zhang N , Torrelles JB , McNeil MR , Escuyer VE , Khoo KH , Brennan PJ , Chatterjee D . The Emb proteins of mycobacteria direct arabinosylation of lipoarabinomannan and arabinogalactan via an N-terminal recognition region and a C-terminal synthetic region , Mol Microbiol , 2003 , vol. 50 (pg. 69 - 76 ) Google Scholar Crossref Search ADS PubMed WorldCat © 2006 The Author(s) This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2006 The Author(s)
TI - Topology and mutational analysis of the single Emb arabinofuranosyltransferase of Corynebacterium glutamicum as a model of Emb proteins of Mycobacterium tuberculosis
JF - Glycobiology
DO - 10.1093/glycob/cwl066
DA - 2007-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/topology-and-mutational-analysis-of-the-single-emb-x01tYyfDjb
SP - 210
EP - 219
VL - 17
IS - 2
DP - DeepDyve
ER -