Role of different moieties from the lipooligosaccharide molecule in biological activities of the Moraxella catarrhalis outer membrane Daxin Peng 1, *, Wei-Gang Hu 1, †, Biswa P. Choudhury 2 , Artur Muszyn ´ ski 2 , Russell W. Carlson 2 and Xin-Xing Gu 1 1 Vaccine Research Section, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Rockville, MD, USA 2 Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA Moraxella catarrhalis, once considered a nonpatho- genic bacteria to humans, is now one of the leading causes of bacterial otitis media in children, after Strep- tococcus pneumoniae and Haemophilus influenzae [1–3]. In developed countries, more than 80% of children under the age of 3 years will be diagnosed at least once Keywords lipooligosaccharide; moiety; Moraxella catarrhalis; outer membrane; role Correspondence Xin-Xing Gu, 5 Research Court, Room 2A31, Rockville, MD 20850, USA Fax: +1 301 435 4040 Tel: +1 301 402 2456 E-mail: guxx@nidcd.nih.gov Present addresses *School of Veterinary Medicine, Yangzhou University, Yangzhou, Jiangsu, China †Defence Research and Development Canada-Suffield, Medicine Hat, Alberta, Canada Database The nucleotide sequence of lgt3 gene in Moraxella catarrhalis strain O35E has been submitted to the GenBank database under the accession number DQ208195 (Received 19 June 2007, revised 30 July 2007, accepted 21 August 2007) doi:10.1111/j.1742-4658.2007.06060.x Lipooligosaccharide (LOS), a major component of the outer membrane of Moraxella catarrhalis, consists of two major moieties: a lipid A and a core oligosaccharide (OS). The core OS can be dissected into a linker and three OS chains. To gain an insight into the biological activities of the LOS molecules of M. catarrhalis, we used a random transposon mutagenesis approach with an LOS specific monoclonal antibody to construct a sero- type A O35E lgt3 LOS mutant. MALDI-TOF-MS of de-O-acylated LOS from the mutant and glycosyl composition, linkage, and NMR analysis of its OS indicated that the LOS contained a truncated core OS and consisted of a Glc-Kdo 2 (linker)-lipid A structure. Phenotypic analysis revealed that the mutant was similar to the wild-type strain in its growth rate, toxicity and susceptibility to hydrophobic reagents. However, the mutant was sensi- tive to bactericidal activity of normal human serum and had a reduced adherence to human epithelial cells. These data, combined with our previ- ous data obtained from mutants which contained only lipid A or lacked LOS, suggest that the complete OS chain moiety of the LOS is important for serum resistance and adherence to epithelial cells, whereas the linker moiety is critical for maintenance of the outer membrane integrity and sta- bility to preserve normal cell growth. Both the lipid A and linker moieties contribute to the LOS toxicity. Abbreviations BHI, brain-heart infusion; CFU, colony forming units; COPD, chronic obstructive pulmonary disease; DPBSG, Dulbecco’s phosphate-buffered saline containing 0.05% gelatin; EU, endotoxin units; Kdo, 3-deoxy- D-manno-2-octulosonic acid; kds, 3-deoxy-D-manno-2-octulosonic acid transferase; LOS, lipooligosaccharide; lpxA, UDP-N-acetylglucosamine acyltransferase; OMPs, outer membrane proteins; OMVs, outer membrane vesicles; OS, oligosaccharide. 5350 FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works with otitis media, and M. catarrhalis is responsible for 15–25% of those cases [4,5]. M. catarrhalis is also a common cause of lower respiratory tract infections in adults with chronic obstructive pulmonary disease (COPD), the fourth leading cause of death in the Uni- ted States [6,7]. Lower respiratory tract infections have been shown to contribute to the progression of the COPD. Approximately 20 million cases of such exacer- bations are reported each year in the United States, up to 35% of them resulting from M. catarrhalis infec- tions. The pathogenesis of M. catarrhalis infection is not well understood. As a gram-negative bacterium, M. catarrhalis without capsular polysaccharides is surrounded by an outer membrane, consisting of lipo- oligosaccharide (LOS), outer membrane proteins (OMPs) and pili outside phospholipids [2]. The outer membrane of gram-negative bacteria was initially con- sidered to protect the bacteria from several antibiotics, dyes, and detergents (hydrophobic compounds) which would normally damage the cell wall. However, more outer membrane biological activities have become apparent over years of extensive research and there is indication today that the outer membrane of gram- negative bacteria has some other biological activities, such as adherence and invasion in vitro, colonization in vivo, maintenance of cell surface charge and serum bactericidal resistance [8–13]. Which component on the outer membrane of M. catarrhalis contributes to the above-mentioned bio- logical activities? Several OMPs of M. catarrhalis were identified, but their functions were mainly reported as potential vaccine antigens, although some of them were indicated to be adhesins [5,14–17]. Pili were con- sidered to be associated with bacterial attachment, but both piliated and nonpiliated M. catarrhalis strains could adhere to human mucosal cells in vitro [18]. LOS has mostly been investigated for these biological activi- ties of the M. catarrhalis outer membrane [19–24]. M. catarrhalis LOS is a major component of the bacterial outer membrane with three main serotypes A, B and C, of which serotype A is the major type [25,26]. Quite a few studies have demonstrated that such an LOS molecule is an important virulence factor for some other respiratory tract pathogens, such as Neisseria meningitidis, Neisseria gonorrhoeae, and H. influenza [27–29]. Studies have also implicated the M. catarrhalis LOS as important in the pathogenesis of the M. catarrhalis infection [13,21,23,24]. We recently reported that a mutant which lacked LOS showed decreased toxicity, reduced resistance to nor- mal human serum killing, and reduced adherence to human epithelial cells in vitro and in vivo, while show- ing hypersensitivity to hydrophobic compounds, indi- cating that LOS contributes to most biological activities of the M. catarrhalis outer membrane [24]. The molecular structure of the M. catarrhalis LOS is smaller than its lipopolysaccharide counterparts in gram-negative pathogens like Escherichia coli and Sal- monella typhimurium and does not have the O-anti- genic side chain of the repeating units characteristic of classical lipopolysaccharide. As showed in Fig. 1, the LOS consists of lipid A and a core oligosaccharide (OS). The latter can be further dissected into three OS chains and a linker composed of a central Glc and two 3-deoxy-d-manno-2-octulosonic acids (Kdo). How do the three moieties play their roles in biological activi- ties of the outer membrane? Biosynthesis of LOSs in M. catarrhalis is complex and strictly sequential, requiring many enzymes such as UDP-glucose-4-epi- merase, Kdo-8-phosphate synthase, Kdo transferase (kdt), glycosyltransferase enzymes, and UDP-N-acetyl- glucosamine acyltransferase (lpxA). Recently, several genes encoding these enzymes have been reported [20,22–24,26,30–32]. A series of stepwise-truncated LOS mutants can therefore be established by knocking out some of the genes and then the biological activities of the different truncated LOS mutants can be Fig. 1. Schematic structure of M. catarrhalis LOS [8]. Lipooligosaccharide (LOS) consists of two major moieties: lipid A and a core oligosaccharide (OS). The core OS can further be dissected into three OS chains and a linker (boxed) composed of a Glc and two Kdo. p, pyranose. D. Peng et al. LOS role in M. catarrhalis outer membrane FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 5351 compared to decipher potential roles of each moiety of the LOS. Previously, we identified an lpxA gene encod- ing UDP-N-acetylglucosamine acyltransferase and a kdtA gene encoding Kdo transferase [23,24]. The lpxA or the kdtA mutant contained no LOS or lipid A-only structure. Both mutants showed a variety of changes of biological activities in vitro as well as in vivo. In this study, we further found an lgt3 gene by a random transposon mutagenesis approach followed with a screen by an LOS-specific mAb. A subsequently iso- genic lgt3 mutant contained a truncated core OS and consisted of a linker (Glc-Kdo 2 )-lipid A structure. The phenotype of this mutant was examined to investigate the roles of different moieties of the LOS molecule in biological activities of the M. catarrhalis outer mem- brane. Results Construction of M. catarrhalis O35Elgt3 mutant M. catarrhalis O35E mutants were constructed by in vitro transposon mutagenesis. The resulting kanamy- cin-resistant colonies were screened for loss of reactiv- ity to a specific anti-LOS mAb 8E7. Two clones were selected and the EZ:TN transposon was found to be inserted at position 901 and 1203 of an open reading frame (ORF) of 1605 bp, respectively (Fig. 2). BLAST searches at GenBank with the deduced polypeptide sequence of the ORF revealed 99.6% and 100% iden- tity with a recently identified lgt3 gene encoding b(1–4) glucosyltransferase of M. catarrhalis serotype B strain 7169 [30] and a putative lgt3 gene from a serotype A strain 25238 [26], respectively. An lgt3 knockout mutant of M. catarrhalis O35E was constructed by allelic exchange with an insertion of a kanamycin-resistant cassette and deletion of an 822 bp fragment between two HindIII sites of the lgt3 (Fig. 2). Silver staining revealed that the mutant LOS mole- cule migrated more rapidly and showed less density than those of the wild-type, indicating a truncated LOS molecule resulting from the mutant (Fig. 3A). Western blot analysis demonstrated that the LOS from O35Elgt3 lost the binding activity to the mAb 8E7 when compared with the wild-type strain (Fig. 3B). Purified LOSs from both strains also showed similar results to those described above (data not shown). To further compare LOS expression levels between the wild-type and mutant strains, the amounts of Kdo from both strains were determined as an indicator. They were 73 ng and 70 ng per 80 lg of outer mem- brane vesicles (OMVs) for the wide type and mutant strains, respectively, indicating the LOS expressions in both strains were similar. Composition analysis by GC-MS showed that the carbohydrates derived from the wild-type O35E strain consisted of Glc, Gal, GlcNAc and Kdo, whereas the carbohydrates derived from the mutant O35Elgt3 con- sisted exclusively of Glc and Kdo. The glycosyl linkage analysis by GC-MS showed that the OSs from the wild-type strain consisted of terminal-, 2-, 4-, and 3,4,6-linked glucosyl residues, terminal- and 4-linked galactosyl residues, and a terminal-linked N-acetylglu- cosaminosyl residue. These linkage results were consis- tent with the structure of the LOS reported for serotype A strain 25238 [33]. The only hexosyl residue in the mutant LOS structure was terminally linked Glc. Fig. 2. Strategy for construction of the lgt3 mutant in M. catarrhalis O35E. Large arrows represent the direction of transcription; the site of the transposon insertion identified in the O35E is denoted as EN:TN, and the location of the deletion replaced by the kanamy- cin-resistant gene (kanR) is between two HindIII cleavage sites. The sites of primers used are indicated as small arrows (25, 26, and 40 of Table 1). AB Fig. 3. LOS patterns of SDS ⁄ PAGE followed by silver staining (A) or western blotting (B) of M. catarrhalis wild-type strain O35E (lane 1) and mutant O35Elgt3 (lane 2). Each lane represents extracts from proteinase K-treated whole cell lysates from 1.9 lg of each bacterial suspension. An anti-LOS MAb 8E7 was used at 1 : 100 dilution (B). Molecular markers (Mark 12; Invitrogen) are indicated on the left. LOS role in M. catarrhalis outer membrane D. Peng et al. 5352 FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works The de-O-acylated LOSs from the wild-type O35E and the mutant O35Elgt3 were further analyzed by MALDI-TOF MS (Fig. 4). The mass spectral data of de-O-acylated LOS from O35E was consistent with the reported structure of serogroup A LOS. Among the major ions m ⁄ z 2672.8(M-H) – corresponds to the molecular mass of the de-O-Ac LOS containing five Glc, two Gal, one GlcNAc, two Kdo and de-O-acyl- ated lipid-A bearing two phosphate moieties. The de- O-acylated lipid-A unit consists of two GlcNAc and two 3-OH C12:0 groups. Among the other ions m ⁄ z of 2654.8, 2694.8 and 2717.4 indicates the presence of the anhydro form and mono- and di-sodiated form of the molecular ion 2672.8 (Fig. 4A). Among other charac- teristic ions, 1558.2 and 1778.4 represent the mass of intact OS with one and two Kdo residues, respectively. The low molecular mass with m ⁄ z of 895.4 is from the Y-type fragment ion from the de-O-acylated lipid-A portion (as observed in the negative mode). The ions with m ⁄ z of 917.4 and 877.4 (unmarked) are mono- sodiated and anhydro forms of 895.4 ion, respectively. The ion with m ⁄ z of 797.4 was from the anhydro frag- ment ion of de-O-acylated lipid-A bearing only one phosphate group (Fig. 4A). The MALDI-TOF mass spectral data of de-O-acyl- ated LOS for the mutant O35Elgt3 showed the pres- ence of ions with m ⁄ z of 1479.6, 1497.6, 1519.6 and 1541.6, respectively. The ion at m ⁄ z 1497.6 (M-H) – corresponds to the deprotonated molecular mass, con- taining one Glc, two Kdo, de-O-acylated lipid-A and two phosphate moieties. The ions with m ⁄ z of 1519.6 and 1541.6 are mono- and di-sodiated forms of the de-O-acylated LOS and the ion with m ⁄ z of 1479.6 is the anhydrous form of the de-O-acylated LOS. The ion 1277.6 is the de-protonated molecular ion from de-O-acylated LOS containing only one Kdo residue (Fig. 4B). The low molecular ions with m ⁄ z of 917.4 and 877.4 are from the mono-sodiated and anhydro forms of the Y-type fragment ion 895.4 arising from the de-O-acylated lipid-A. Finally, proton NMR spectra of carbohydrates from both wild and mutant strains showed that the spec- trum for the O35E carbohydrates was identical to that published from strain 25238 LOS (Fig. 5) [33]. The mutant carbohydrate spectrum showed only one ano- meric resonance, a result consistent with a single termi- nal-linked glucosyl residue in the above GC-MS analysis. Taken together, the O35Elgt3 mutant contained only a linker-lipid A (Glc-Kdo 2 -lipid A) structure without OS chains (Fig. 1). Morphology, growth rate, and OMV ⁄ OMP profiles of M. catarrhalis O35Elgt3 mutant There were no significant differences in morphology, growth rate, or OMV ⁄ OMP profiles between the wild- type and the lgt3 mutant (data not shown). The yields of OMVs from both strains were similar (data not shown). Susceptibility of M. catarrhalis O35Elgt3 mutant A broad range of hydrophobic agents and a hydrophilic glycopeptide were used to determine the susceptibility of the mutant O35Elgt3. O35Elgt3 showed similar resis- tance or intermediate susceptibility to hydrophobic anti- biotics, reagents [azithromycin (15 lg), deoxycholate (100 mgÆmL )1 ), fusidic acid (10 mgÆmL )1 ), novobiocin A B Fig. 4. MALDI-TOF MS spectrum (negative mode) of the de-O-acyl- ated LOS from M. catarrhalis wild-type strain O35E (A) and the mutant O35Elgt3 (B). This analysis was done in the negative mode, and all ions represented as deprotonated [M-H] – ions. The source of the ions was as indicated in the structure shown in the inset. The m ⁄ z 917.4 and 797.4 are due to sodiated and dehydrated forms, respectively, of the m ⁄ z 895.4 fragment ion. D. Peng et al. LOS role in M. catarrhalis outer membrane FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 5353 (5 lg), polymycin B (300 international units), rifampin (5 lg), Triton X-100 (5%, w ⁄ v), Tween 20 (5%, v ⁄ v)] and a hydrophilic glycopeptide, vancomycin (5 lg), as the wild-type strain (data not shown). Biological activity of M. catarrhalis O35Elgt3 mutant The mutant was tested for LOS-associated biological activity. In a Limulus amebocyte lysate assay, whole cell suspensions (D 620 ¼ 0.1) gave 3.7 · 10 3 endotoxin units (EU) mL )1 for O35E and 2.0 · 10 3 EUÆmL )1 for the O35Elgt3 mutant. In a bactericidal assay with normal human serum, strain O35E survived at 25% normal human serum. However, 75% of the mutant cells died at 25% normal human serum (P<0.05, Fig. 6) indicating a reduced resistance from the mutant. To test the adherence of the O35Elgt3 mutant to human epithelial cells, Chang and HeLa cell lines were used. The adherence percentage of O35Elgt3 to Chang and HeLa epithelia were 19.1 ± 4.6 and 27.3 ± 8.4, respectively, whereas those of the wild-type were 41.7 ± 9.1 and 47.0 ± 4.1 (P<0.01). To investigate the effect on survival of the O35Elgt3 mutant in a murine model of nasopharyngeal clear- ance, mice were challenged with the wild-type or mutant strains by aerosolization (Fig. 7). Mutant A B Fig. 5. Proton NMR spectra of LOS derived from M. catarrhalis O35E (A) and the mutant O35Elgt3 (B). For the O35E strain, the assignments were made based on the identity of this spectrum with that of the published structure for the LOS from type A M. catarrhalis. The spectrum for the mutant lgt3 disaccharide shows numerous resonances for the H3 protons of Kdo due to the fact that a 5-linked Kdo residue can, on mild acid hydrolysis, form 4,8-anhydroKdo, 4,7-anhydroKdo, and 2,7-anhydroKdo; all of which have different H3 resonances. 5.5 6 6.5 7 0 0.5 2.5 5 12.5 25 HI Normal human Serum (%) Bacterial counts (log CFU) * Fig. 6. Bactericidal activity of normal human serum against M. catarrhalis wild-type strain O35E (black bar) and the mutant O35Elgt3 (gray bar). ‘HI’ represents the group of 25% heat- inactivated normal human serum. The data represent the average of three independent assays. An asterisk represents statistical sig- nificance between the wild-type and mutant strains. 0 1 2 3 4 5 0 Nasal wash * * Time p ost challen g e (h) Bacterial recovery (log CFU/mouse) 3624 Fig. 7. Time courses of bacterial recovery in mouse nasal washes after an aerosol challenge with M. catarrhalis wild-type strain O35E (r) and the mutant O35Elgt3 (h). Each time point represents a geometric mean of six mice. Asterisks represent statistical signifi- cance between the wild-type and mutant strains. LOS role in M. catarrhalis outer membrane D. Peng et al. 5354 FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works O35Elgt3 present in the nasopharynx showed an accel- erated clearance at 3 h (85.5% versus 60.2%, P < 0.01) or 6 h (98.0% versus 86.8%, P < 0.01) compared to the wild-type. Discussion In this study, an lgt3 gene was found and its products confirmed from a M. catarrhalis serotype A strain O35E through an isogenic O35Elgt3 mutant. Structural analysis revealed that the O35Elgt3 mutant contained a truncated core OS and consisted of a linker (Glc- Kdo 2 )-lipid A structure, similar to that reported previ- ously [30,34]. The O35Elgt3 mutant demonstrated no significant difference in its growth rate, LOS expres- sion level and susceptibility to the hydrophobic com- pounds compared with the wild-type strain. However, our previous mutants which lacked LOS [24] or con- tained only lipid-A structure [23] showed reduced growth rate and hypersusceptibility to the hydrophobic compounds. It implies that the linker moiety of the LOS might be critical for maintenance of outer mem- brane integrity, stability or flexibility by resistance to foreign hydrophobic compounds to preserve normal cell growth. In addition, the O35Elgt3 mutant was sen- sitive to the bactericidal activity of normal human serum at 25% as compared with the wild-type strain, but less sensitive than that of the mutant with lipid A- only structure [23]. It indicates that the linker moiety (Glc-Kdo 2 ) might also contribute to the serum bacteri- cidal resistance. However, another possibility also exists. As the susceptibility of the lipid A-only mutant to hydrophobic compounds was higher than that of the O35Elgt3 mutant, the permeability barrier of the outer membrane might significantly contribute to the bactericidal activity of normal human serum, and might cause a much more sensitive phenotype of the lipid A-only mutant than that of the O35Elgt3 mutant. Nevertheless, as the O35Elgt3 mutant did not change its susceptibility to hydrophobic compounds but was shown to be more sensitive to the bactericidal activity of the human serum as compared with the wild-type strain, it suggests that a complete OS chain moiety contributes to the serum bactericidal resistance. LOS toxicity was assumed to be associated only with the lipid A moiety. By using Limulus amebocyte lysate assay, we found that the O35E lgt3 mutant did not show significant changes of the toxicity, whereas the lipid A-only mutant [23] showed decreased toxicity (6 · 10 2 EU Æ mL )1 ) six-fold as compared with the wild-type strain (3.7 · 10 3 EU Æ mL )1 ), suggesting that the LOS toxicity is attributable not only to the lipid A moiety but also to the linker moiety (Glc-Kdo 2 ). Bacterial adherence to the surface of epithelial cells plays a critical role in colonization and is believed to be the first step in the pathogenesis of microbial infec- tions. It has been reported that several LOSs from respiratory tract bacteria are associated with bacterial adherence [29,35,36]. Consistent with our previous results from the lipid A-only mutant, the O35Elgt3 mutant showed more than a 50% reduction in attach- ment to Chang or HeLa cells and enhanced clearance from the mouse nasopharynx after an aerosol chal- lenge when compared to the wild-type strain [23]. Our data suggest that the M. catarrhalis adherence to epi- thelial cells may not be associated with the linker and lipid A moieties, but with the OS chain moiety. In summary, we constructed a serotype A M. catarrhalis O35Elgt3 mutant that produced a linker (Glc-Kdo 2 )-lipid A structure. The mutant showed sig- nificant changes in its biological activities in vitro and in vivo. These data, combined with our previous data from mutants which lacked LOS or contained only lipid A structure, suggest that a complete OS chain moiety of the M. catarrhalis LOS is important in bio- logical activities such as serum resistance and adher- ence of epithelial cells, the linker moiety is critical for maintenance of outer membrane integrity and stability to preserve normal cell growth, and both lipid A and linker moieties contribute to the LOS toxicity. Experimental procedures Strains, plasmids, and growth conditions Bacterial strains, plasmids and primers are given in Table 1. M. catarrhalis strains were cultured on chocolate agar plates (Remel, Lenexa, KS), or brain-heart infusion (BHI) (Difco, Detroit, MI) agar plates at 37 °Cin5%CO 2 . Mutant strains were selected on BHI agar supplemented with kanamycin at 20 lgÆmL )1 . Growth rates of wild-type and mutant were measured as follows: an overnight culture was inoculated in 10 mL of BHI media (adjusted D 600 ¼ 0.05) and shaken at 37 °C at 250 r.p.m. The bacterial cultures were monitored spectrophotometrically at D 600 . Escherichia coli was grown on Luria-Bertani agar plates or broth with appropriate antibiotic supplementation. The antibiotic concentrations used for E. coli were as follows: kanamycin, 30 lgÆmL )1 ; and ampicillin, 50 lgÆmL )1 . General DNA methods DNA restriction endonucleases, T4 DNA ligase, E. coli DNA polymerase I Klenow fragment, and Taq DNA poly- merase were purchased from Fermentas (Hanover, MD). Preparation of plasmids, purification of PCR products and D. Peng et al. LOS role in M. catarrhalis outer membrane FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 5355 DNA fragments were performed using kits manufactured by Qiagen (Santa Clarita, CA). Bacterial chromosomal DNA was isolated using a genomic DNA purification kit (Promega, Madison, WI). DNA nucleotide sequences were obtained via 3070xl DNA analyzer (Applied Biosystems, Foster City, CA) and analyzed with dnastar software (DNASTAR Inc., Madison, WI). Transposon mutagenesis and identification of lgt3 gene In vitro transposon mutagenesis of M. catarrhalis was per- formed as described previously [23]. The resulting kana- mycin-resistant colonies were screened by colony blot assay and further identified by whole cell enzyme-linked immunosorbent assay (ELISA) for loss of binding reactiv- ity to an anti-LOS MAb (8E7) generated by serotype A strain O35E [21]. The transformants without the 8E7 binding activity were subsequently evaluated by examin- ing the LOS profiles from proteinase-K-treated whole cell lysates [37] on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS ⁄ PAGE) and LOS silver staining [38]. The clones with different LOS bands from the wild-type LOS were selected for isolation of chromosomal DNA. The DNA fragment with the transposon insertion was subsequently sequenced and the lgt3 homologue from strain O35E identified by BLAST searches at GenBank of the National Center for Biotechnology Information, Bethesda, MD. Cloning of lgt3 homologue and construction of the knockout mutants The entire lgt3 was amplified from the chromosomal DNA of M. catarrhalis strain O35E using primers 25 and 26 (Table 1, Fig. 2). The PCR product was cloned into pCR2.1 using a TOPO TA cloning kit (Invitrogen, Carls- bad, CA) to obtain pCRlgt3. The insert was released by EcoRI-SalI digestion and then subcloned into an EcoRI- SalI site of pBluescript SK(+) to form pSlgt3. The kana- mycin-resistant gene from pUC4K was digested with EcoRI and subsequently inserted into the lgt3 gene using the HindIII site to form pSlgt3K. After verification by sequence analysis, the mutagenic lgt3 gene with the inserted kanamycin-resistant gene in pSlgt3K was amplified by PCR and purified for electropo- ration to O35E-competent cells. After 24 h incubation, the resulting kanamycin-resistant colonies were selected for PCR analysis of chromosomal DNA using primers 40 and 26 (Table 1, Fig. 2). An inactivated lgt3 mutant was verified by sequence analysis and named as O35Elgt3. LOS determination A crude LOS extraction was performed from both the wild- type strain O35E and the mutant O35Elgt3 using the proteinase-K-treated whole cell lysate method [37]. The resulting extracts from each bacterial suspension (1.9 lgof protein amount) were resolved by 15% SDS ⁄ PAGE and Table 1. Strains, plasmids, and primers used in this study. Strains Description Source E. coli TOP10 Cloning strain Invitrogen M. catarrhalis O35E Wild-type strain (24) M. catarrhalis O35E7 Strain with EZ::TN insertion in lgt3 gene (1203-bp) This study M. catarrhalis O35E8 Strain with EZ::TN insertion in lgt3 gene (901-bp) This study M. catarrhalis O35Elgt3 lgt3 knockout mutant strain of O35E This study Plasmids pCR2.1 TOPO TA cloning vector Invitrogen pBluescript II SK(+) Cloning vector Fermentas pUC4k Kanamycin-resistant gene Amersham pCRlgt3 lgt3 cloned into pCR2.1 This study pSlgt3 EcoRI-SalI lgt3 fragment cloned into SK(+) This study pSlgt3K EcoRI-blunted kanamycin-resistant gene inserted into blunted HindIII site of pSlgt3 with internal deletion This study Primers 25 5¢-CTC GTC GAC ATG AAT ACA GCC AAG CGG-3¢ This study lgt3 sense, SalI site underlined 26 5¢-CTC GAA TTC TCA TGG TTT ATC CTT ATT-3¢ This study lgt3 antisense, EcoRI site underlined 40 5¢-CAA CCC ATT TTC TAA GCT-3¢ This study 5¢ flanking DNA of lgt3 sense LOS role in M. catarrhalis outer membrane D. Peng et al. 5356 FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works visualized by silver staining [38]. Western blot using MAb 8E7 was performed [21]. To further quantify LOSs expressed on both the wild- type and mutant strains, a Kdo assay was applied [39] with OMVs prepared using an EDTA-heat induced method [40] as whole cells were not applicable for the Kdo assay. The amount of Kdo was determined by using 80 lg of OMVs as samples and Kdo ammonium salt (Sigma, St. Louis, MO) as a standard. Structural analysis of LOS For the composition analysis, 30–35 g of wet cells from O35E and O35Elgt3 were prepared for LOS purification by phenol–water extraction [41]. The purified LOSs were washed with a 9 : 1 ethanol ⁄ water (v ⁄ v) mixture to remove phospholipids. The OSs were prepared by mild acid hydro- lysis of LOSs in 1% aqueous acetic acid (v ⁄ v) for 2.5 h at 100 °C and gel-filtration chromatography using Bio-Gel P-2 with deionized water as the eluent. The glycosyl composi- tions of OSs in O35E and O35Elgt3 were determined by the preparation and GC-MS analysis of trimethylsilyl methyl- glycosides. The glycosyl linkages were determined by the preparation and GC-MS analysis of partially methylated aldiol acetates. Analysis of GC-MS was performed on an HP)5890 GC interfaced to a mass selective detector 5970 MSD using a Supelco DB1 fused silica capillary col- umn (30 m · 0.25 mm Internal diameter, J & W Scientific, Folsom, California). For MS analysis, the LOS was O-deacylated by treatment with anhydrous hydrazine for 20 min at 37 °C [42]. The de-O-acylated LOS samples were dissolved in deionized water at 1 l g ÆlL )1 concentration and mixed with equal volumes of 0.5 m 2,5-dihydroxy benzoic acid (matrix) in methanol and spotted with 1 lL on a 100- well stainless steel MALDI plate. The mass spectra were col- lected on a MALDI-TOF instrument (Applied Biosystems) in the negative reflectron mode. NMR was performed on the OSs by lyophilizing each sample from D 2 O (99.999 atom percentage D, Sigma) twice, dissolved in D 2 O (100 atom percentage D), and acquiring spectra using a Varian Ino- va 500 or 600 MHz spectrometer (Varian, Palo Alto, CA). Limulus amebocyte lysate assay The chromogenic Limulus amebocyte lysate assay for endotoxin activity was performed using the QCL-1000 kit (Bio-Whittaker Inc., Walkersville, MD). Overnight cultures from chocolate agar plates were suspended in BHI broth to D 620 of 0.1 and serial dilutions of these stocks were tested. Susceptibility determination The sensitivity of strains to a panel of hydrophobic agents or a hydrophilic glycopeptide was performed using standard disk-diffusion assays [43]. Bacteria were cultured in BHI to a D 600 of 0.2 and 100 lL portions of the bacte- rial suspension were spread onto chocolate agar plates. Antibiotic disks or sterile blank paper disks (6 mm, Becton Dickinson, Cockeysville, MD) saturated with the various agents were plated on the lawn in triplicate at 37 °C for 18 h. Sensitivity was assessed by measuring the diameter of the zone of growth inhibition in two axes and the mean value was calculated. Bactericidal assay to normal human serum A complement-sufficient normal human serum was pre- pared and pooled from eight healthy adult donors. A bacte- ricidal assay was performed in a 96-well plate [23]. Normal human serum was diluted to 0.5, 2.5, 5.0, 12.5, and 25% in pH 7.4 Dulbecco’s phosphate-buffered saline (NaCl ⁄ P i ) containing 0.05% gelatin (DPBSG). Bacteria (10 lLof10 6 colony forming units, CFU) were inoculated into 190 lL reaction wells containing the diluted normal human serum, 25% of heat-inactivated normal human serum, or DPBSG alone and incubated at 37 °C for 30 min. Serial dilutions (1 : 10) of each well were plated onto chocolate agar plates. The resulting colonies were counted after 24 h of incuba- tion. Adherence assay Chang (conjunctival; CCL20.2) and HeLa (cervix; CCL-2) human epithelial lines were cultured in Eagle’s minimal essential medium (ATCC, Manassas, VA) supplemented with 10% heat-inactivated fetal bovine serum in an atmo- sphere of 5% CO 2 at 37 °C. A quantitative adherence assay was performed on a 24-well tissue culture plate (Corning Incorporated, Corning, NY) [34]. Adherence was expressed as the percentage of bacteria attached to the human cells relative to the original bacteria added to the well. The data represent the average of three independent assays. Nasopharyngeal clearance pattern in animal model Female BALB ⁄ c mice (6–8 weeks of age) were obtained from Taconic Farms Inc. (Germantown, NY). The mice were housed in an animal facility in accordance with National Institutes of Health guidelines under animal study protocol 1158–04. Bacterial aerosol challenges were carried out in mice using the same D 540 value for wild-type strain O35E (1.24 · 10 9 CFUÆmL )1 ) or mutant O35Elgt3 (8.2 · 10 8 CFUÆmL )1 ) in a 10 mL DPBSG [44]. The num- bers of bacteria present in nasal washes were measured at various time points postchallenge. The minimum detectable number of viable bacteria was 4 CFU per nasal washing. Clearance of M. catarrhalis was expressed as the percentage D. Peng et al. LOS role in M. catarrhalis outer membrane FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 5357 of bacterial CFU at each time point compared with the number deposited at time zero. Statistical analysis The adherence and clearance percentages were analyzed by a Chi-square test. Acknowledgements We thank Eric J. Hansen for providing strain O35E, Wenzhou Hong for assisting in the animal challenge, and Robert Morell and Yandan Yang for helping in DNA sequencing. This research was supported by the Intramural Research Program of the NIH, NIDCD. The analytical work was supported in part by the Department of Energy-funded (DE-FG09–93ER20097) Center for Plant and Microbial Complex Carbohy- drates. References 1 Catlin BW (1990) Branhamella catarrhalis: an organism gaining respect as a pathogen. Clin Microbiol Rev 3, 293–320. 2 Murphy TF (1996) Branhamella catarrhalis: epidemiol- ogy, surface antigenic structure, and immune response. Microbiol Rev 60, 267–279. 3 Karalus R & Campagnari A (2000) Moraxella catarrh- alis: a review of an important human mucosal pathogen. Microbes Infect 2, 547–559. 4 Ruuskanen O & Heikkinen T (1994) Otitis media: etiol- ogy and diagnosis. Pediatr Infect Dis J 13, S23–S26; dis- cussion S50–S54. 5 Faden H, Hong J & Murphy T (1992) Immune response to outer membrane antigens of Moraxella catarrhalis in children with otitis media. Infect Immun 60, 3824–3829. 6 Verduin CM, Hol C, Fleer A, van Dijk H & van Belkum A (2002) Moraxella catarrhalis: from emerging to established pathogen. Clin Microbiol Rev 15, 125–144. 7 Murphy TF, Brauer AL, Grant BJ & Sethi S (2005) Moraxella catarrhalis in chronic obstructive pulmonary disease: burden of disease and immune response. Am J Respir Crit Care Med 172, 195–199. 8 Holme T, Rahman M, Jansson PE & Widmalm G (1999) The lipopolysaccharide of Moraxella catarrhalis structural relationships and antigenic properties. Eur J Biochem 265, 524–529. 9 Weiser JN & Gotschlich EC (1991) Outer membrane protein A (OmpA) contributes to serum resistance and pathogenicity of Escherichia coli K-1. Infect Immun 59, 2252–2258. 10 Hill SA (2000) Opa expression correlates with elevated transformation rates in Neisseria gonorrhoeae. J Bacte- riol 182, 171–178. 11 de O Ferreira E, Araujo Lobo L, Barreiros Petropolis D, Dos S Avelar KE, Ferreira MC, e Silva Filho FC, Domingues RM (2006) A Bacteroides fragilis surface glycoprotein mediates the interaction between the bacte- rium and the extracellular matrix component laminin-1. Res Microbiol 157, 960–966. 12 Shin S, Lu G, Cai M & Kim KS (2005) Escherichia coli outer membrane protein A adheres to human brain microvascular endothelial cells. Biochem Biophys Res Commun 330 , 1199–1204. 13 Attia AS, Lafontaine ER, Latimer JL, Aebi C, Syro- giannopoulos GA & Hansen EJ (2005) The UspA2 pro- tein of Moraxella catarrhalis is directly involved in the expression of serum resistance. Infect Immun 73, 2400– 2410. 14 Easton DM, Smith A, Gallego SG, Foxwell AR, Cripps AW & Kyd JM (2005) Characterization of a novel porin protein from Moraxella catarrhalis and identifica- tion of an immunodominant surface loop. J Bacteriol 187, 6528–6535. 15 McMichael JC (2000) Vaccines for Moraxella catarrhal- is. Vaccine 19 (Suppl. 1), S101–S107. 16 Murphy TF, Kirkham C, Liu DF & Sethi S (2003) Human immune response to outer membrane protein CD of Moraxella catarrhalis in adults with chronic obstructive pulmonary disease. Infect Immun 71, 1288– 1294. 17 Pearson MM, Laurence CA, Guinn SE & Hansen EJ (2006) Biofilm formation by Moraxella catarrhalis in vitro: roles of the UspA1 adhesin and the Hag hemagglutinin. Infect Immun 74, 1588–1596. 18 Ahmed K, Matsumoto K, Rikitomi N & Nagatake T (1996) Attachment of Moraxella catarrhalis to pharyn- geal epithelial cells is mediated by a glycosphingolipid receptor. FEMS Microbiol Lett 135, 305–309. 19 Akgul G, Erturk A, Turkoz M, Turan T, Ichinose A, Nagatake T & Ahmed K (2005) Role of lipooligosac- charide in the attachment of Moraxella catarrhalis to human pharyngeal epithelial cells. Microbiol Immunol 49, 931–935. 20 Zaleski A, Scheffler NK, Densen P, Lee FK, Campag- nari AA, Gibson BW & Apicella MA (2000) Lipooligo- saccharide P (k) (Galalpha1–4Galbeta1–4Glc) epitope of Moraxella catarrhalis is a factor in resistance to bacteri- cidal activity mediated by normal human serum. Infect Immun 68 , 5261–5268. 21 Hu WG, Chen J, McMichael JC & Gu XX (2001) Functional characteristics of a protective monoclonal antibody against serotype A and C lipooligosaccharides from Moraxella catarrhalis. Infect Immun 69, 1358– 1363. LOS role in M. catarrhalis outer membrane D. Peng et al. 5358 FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 22 Luke NR, Allen S, Gibson BW & Campagnari AA (2003) Identification of a 3-deoxy-D-manno-octulosonic acid biosynthetic operon in Moraxella catarrhalis and analysis of a KdsA-deficient isogenic mutant. Infect Immun 71, 6426–6434. 23 Peng D, Choudhury BP, Petralia RS, Carlson RW & Gu XX (2005) Roles of 3-deoxy-D-manno-2-octulosonic acid transferase from Moraxella catarrhalis in lipooligo- saccharide biosynthesis and virulence. Infect Immun 73, 4222–4230. 24 Peng D, Hong W, Choudhury BP, Carlson RW & Gu XX (2005) Moraxella catarrhalis bacterium without endotoxin, a potential vaccine candidate. Infect Immun 73, 7569–7577. 25 Vaneechoutte M, Verschraegen G, Claeys G & Van Den Abeele AM (1990) Serological typing of Branham- ella catarrhalis strains on the basis of lipopolysaccharide antigens. J Clin Microbiol 28, 182–187. 26 Edwards KJ, Schwingel JM, Datta AK & Campagnari AA (2005) Multiplex PCR assay that identifies the major lipooligosaccharide serotype expressed by Morax- ella catarrhalis clinical isolates. J Clin Microbiol 43, 6139–6143. 27 Song W, Ma L, Chen R & Stein DC (2000) Role of lipooligosaccharide in Opa-independent invasion of Neisseria gonorrhoeae into human epithelial cells. J Exp Med 191, 949–960. 28 Gorter AD, Oostrik J, van der Ley P, Hiemstra PS, Dankert J & van Alphen L (2003) Involvement of lipooligosaccharides of Haemophilus influenzae and Neisseria meningitidis in defensin-enhanced bacterial adherence to epithelial cells. Microb Pathog 34, 121– 130. 29 Swords WE, Chance DL, Cohn LA, Shao J, Apicella MA & Smith AL (2002) Acylation of the lipooligosac- charide of Haemophilus influenzae and colonization: an htrB mutation diminishes the colonization of human airway epithelial cells. Infect Immun 70, 4661–4668. 30 Edwards KJ, Allen S, Gibson BW & Campagnari AA (2005) Characterization of a cluster of three glycosyl- transferase enzymes essential for Moraxella catarrhalis lipooligosaccharide assembly. J Bacteriol 187, 2939– 2947. 31 Peak IR, Grice ID, Faglin I, Klipic Z, Collins PM, van Schendel L, Hitchen PG, Morris HR, Dell A & Wilson JC (2007) Towards understanding the functional role of the glycosyltransferases involved in the biosynthesis of Moraxella catarrhalis lipooligosaccharide. FEBS J 274, 2024–2037. 32 Wilson JC, Collins PM, Klipic Z, Grice ID & Peak IR (2006) Identification of a novel glycosyltransferase involved in LOS biosynthesis of Moraxella catarrhalis. Carbohydr Res 341, 2600–2606. 33 Edebrink P, Jansson PE, Rahman MM, Widmalm G, Holme T, Rahman M & Weintraub A (1994) Structural studies of the O-polysaccharide from the lipopolysac- charide of Moraxella (Branhamella) catarrhalis serotype A (strain ATCC 25238). Carbohydr Res 257, 269–284. 34 Aebi C, Lafontaine ER, Cope LD, Latimer JL, Lumb- ley SL, McCracken GH Jr & Hansen EJ (1998) Pheno- typic effect of isogenic uspA1 and uspA2 mutations on Moraxella catarrhalis 035E. Infect Immun 66, 3113– 3119. 35 Albiger B, Johansson L & Jonsson AB (2003) Lipooli- gosaccharide-deficient Neisseria meningitidis shows altered pilus-associated characteristics. Infect Immun 71, 155–162. 36 Swords WE, Buscher BA, Ver Steeg Ii K, Preston A, Nichols WA, Weiser JN, Gibson BW & Apicella MA (2000) Non-typeable Haemophilus influenzae adhere to and invade human bronchial epithelial cells via an inter- action of lipooligosaccharide with the PAF receptor. Mol Microbiol 37, 13–27. 37 Tzeng YL, Datta A, Kolli VK, Carlson RW & Stephens DS (2002) Endotoxin of Neisseria meningitidis com- posed only of intact lipid A: inactivation of the menin- gococcal 3-deoxy-D-manno-octulosonic acid transferase J Bacteriol 184, 2379–2388. 38 Tsai CM & Frasch CE (1982) A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal Biochem 119, 115–119. 39 Weissbach A & Hurwitz J (1959) The formation of 2-keto-3-deoxyheptonic acid in extracts of Escherichia coli B. I. Identification. J Biol Chem 234 , 705–709. 40 Murphy TF & Loeb MR (1989) Isolation of the outer membrane of Branhamella catarrhalis. Microb Pathog 6, 159–174. 41 Gu XX, Chen J, Barenkamp SJ, Robbins JB, Tsai CM, Lim DJ & Battey J (1998) Synthesis and characteriza- tion of lipooligosaccharide-based conjugates as vaccine candidates for Moraxella (Branhamella) catarrhalis. Infect Immun 66, 1891–1897. 42 Helander IM, Lindner B, Brade H, Altmann K, Lind- berg AA, Rietschel ET & Zahringer U (1988) Chemical structure of the lipopolysaccharide of Haemophilus influenzae strain I-69 Rd- ⁄ b+. Description of a novel deep-rough chemotype. Eur J Biochem 177, 483–492. 43 Ochsner UA, Vasil AI, Johnson Z & Vasil ML (1999) Pseudomonas aeruginosa fur overlaps with a gene encoding a novel outer membrane lipoprotein, OmlA. J Bacteriol 181, 1099–1109. 44 Hu WG, Chen J, Battey JF & Gu XX (2000) Enhance- ment of clearance of bacteria from murine lungs by immunization with detoxified lipooligosaccharide from Moraxella catarrhalis conjugated to proteins. Infect Immun 68, 4980–4985. D. Peng et al. LOS role in M. catarrhalis outer membrane FEBS Journal 274 (2007) 5350–5359 Journal compilation ª 2007 FEBS. No claim to original US government works 5359 . Role of different moieties from the lipooligosaccharide molecule in biological activities of the Moraxella catarrhalis outer membrane Daxin Peng 1, *,. (OS). The core OS can be dissected into a linker and three OS chains. To gain an insight into the biological activities of the LOS molecules of M. catarrhalis,