Novel Sialic Acid Transporter of Haemophilus influenzae
http://www.100md.com
感染与免疫杂志 2005年第9期
Buck Institute for Age Research, 8001 Redwood Blvd., Novato, California 94945
Department of Microbiology, University of Iowa, Iowa City, Iowa 52242
Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94143-0446
ABSTRACT
Nontypeable Haemophilus influenzae is an opportunistic pathogen and a common cause of otitis media in children and of chronic bronchitis and pneumonia in patients with chronic obstructive pulmonary disease. The lipooligosaccharides, a major component of the outer membrane of H. influenzae, play an important role in microbial virulence and pathogenicity. N-Acetylneuraminic acid (sialic acid) can be incorporated into the lipooligosaccharides as a terminal nonreducing sugar. Although much of the pathway of sialic acid incorporation into lipooligosaccharides is understood, the transporter responsible for N-acetylneuraminic acid uptake in H. influenzae has yet to be characterized. In this paper we demonstrate that this transporter is a novel sugar transporter of the tripartite ATP-independent periplasmic transporter family. In the absence of this transporter, H. influenzae cannot incorporate sialic acid into its lipooligosaccharides, making the organism unable to survive when exposed to human serum and causing reduced viability in biofilm growth.
INTRODUCTION
Haemophilus influenzae is exclusively adapted to infect or colonize humans. Strains can be encapsulated or nonencapsulated (nontypeable). Nontypeable Haemophilus influenzae (NTHi) is a frequent colonizer of the nasopharynx. When the airway is compromised, NTHi can cause local infections such as otitis media in young children (24 million physician visits per year in the United States [25]) and chronic bronchitis and pneumonia in patients with chronic obstructive pulmonary disease.
Lipooligosaccharides (LOS) are a major component of the NTHi outer membrane and have been shown to play a role in microbial virulence and pathogenicity (22). LOS contains carbohydrate epitopes which mimic human glycosphingolipids, allowing the bacteria to avoid the host immune response (17). LOS present on the surface of NTHi is a heterogeneous mixture of glycoforms, the most abundant of which has been extensively studied and is known to consist of a lactose moiety (Gal1-4Glc) attached to the first heptose (HepI) of a conserved core structure [HepIII1,2-HepII1,3-HepI1,5-Kdo(P)-lipid A] (19, 20, 24). Importantly, NTHi is also capable of incorporating the acidic sugar N-acetylneuraminic acid (Neu5Ac, or sialic acid) as terminal nonreducing units into its LOS, giving the bacterium protection from complement-mediated killing by normal human serum (10, 11). The acceptors for sialic acid are lactose (18), N-acetyllactosamine, and possibly N-acetylgalactosamine, although the precise structures of most of these sialylated LOS species have not been conclusively identified (12).
The sialic acid is incorporated into the LOS before it reaches the cell surface by one of three sialyltransferases, SiaA, Lic3a, or LsgB (10, 12). The donor for this transfer is CMP-sialic acid, which is synthesized from sialic acid and CTP by the CMP-sialic acid synthetase (SiaB) (11). The fate of sialic acid in NTHi is not solely incorporation into the LOS; sialic acid can also be utilized as a carbon and nitrogen source via its breakdown to N-acetylmannosamine and pyruvate by the neuraminyl lyase (NanA) (28).
NTHi is incapable of synthesizing sialic acid and thus requires an exogenous source of sialic acid for incorporation to occur. In Escherichia coli, sialic acid is imported via symport with a proton through a specific transporter (NanT) of the major facilitator superfamily (30). A gene (HI1104) was identified in the H. influenzae genome that has high homology to the E. coli sialic acid transporter. HI1104 was deleted in this study and shown to have no effect on sialic acid uptake in the H. influenzae strain studied here. Recent publications have suggested that sialic acid transport in H. influenzae is mediated via a novel class of transporter, a tripartite ATP-independent periplasmic (TRAP) transporter (15, 29). TRAP transporters consist of three components: an extracellular solute receptor (ESR) and two distinct integral membrane components of unequal size which are sometimes fused (14). These transporters differ from the better-characterized ABC-protein transporter family (4) in that they do not possess an ATP-binding cassette protein and are not driven by ATP hydrolysis but rather by an electrochemical ion gradient (14).
The gene HI0147 (siaT) was previously identified by Rabus et al. (21) as the fused transmembrane domains of a TRAP transporter and named Y147; it was recently suggested to be part of a sialic acid transporter (15, 29). In this paper we provide evidence that the siaT gene product is indeed a component of the sialic acid TRAP transporter in the NTHi strain 2019. Deletion of the gene encoding this protein has a marked effect on the incorporation of sialic acid into the LOS and on the survival of the organism when exposed to human serum.
MATERIALS AND METHODS
Bacterial growth. Strains and plasmids used in this study are listed in Table 1. Escherichia coli was grown at 37°C in Luria-Bertani medium with or without agar (1.5%) and supplemented with antibiotics as needed. Wild-type H. influenzae was grown on supplemented brain heart infusion (BHI) agar (Difco Laboratories, Detroit, MI) supplemented with 10 μg/ml hemin and 10 μg/ml NAD at 37°C. Erythromycin-resistant H. influenzae was selected on supplemented BHI agar with 5 μg of erythromycin/ml. Selection was carried out without CO2.
Cloning and mutagenesis of HI1104. PCR was used to amplify a 2,146-bp fragment from NTHi 2019 genomic DNA containing the HI1104 open reading frame (ORF) by using oligonucleotide primers (5'-TCCCCCCGGGTCATGGAAAGATACGGATGCAAAG-3' and 5'-TCCCCCCGGGTCAAAAGGCGACAAAGAGGGTGG-3') with restriction sites for SmaI (underlined). This fragment was digested with SmaI and cloned into the SmaI site in pACYC177 (New England Biolabs, Beverly, MA). The sequence of the fragment was confirmed by sequencing and comparison with the H. influenzae Rd Kw-20 genome. This construct was named p20191104. The SmaI fragment of pBSLerm containing the erythromycin resistance gene (mlsR) was inserted into StyI/BsaBI-digested and blunt end-filled p20191104 by eliminating 759 of 1,220 bp of the 2019 HI1104 ORF. This plasmid was designated p20191104erm. NTHi 2019 was transformed with the 2,882-bp SmaI fragment of p20191104erm that was isolated away from the plasmid backbone by electrophoresis in an agarose gel. The NTHi mutant was confirmed by PCR and Southern blot analysis.
Cloning and mutation of NTHi 2019siaT. A 1,762-base-pair DNA fragment was amplified from genomic DNA of NTHi strains 2019, 3198, and 7502 by PCR with primers 147-up (5'-TTTCCTACACGAGCAACAAC-3') and 147-down (5'-CTACATTCCCTTATTCTTCATCAAAC-3'). This fragment was cloned by ligation into the pCR2.1TOPO vector, and transformation of DH5 host bacteria was carried out using the manufacturer's protocols (Invitrogen, Carlsbad, CA). The sequences of the TA inserts were determined. These fragments corresponded to bases 16 to 1778 of the 1,902-base-pair HI0147 open reading frame from the complete genome of H. influenzae Rd KW-20 (www.tigr.org). These plasmids were named p2019HI0147, p3198HI0147, and p7502HI0147.
Only p7502HI0147 had a convenient restriction enzyme site near the center of the insert sequence. The SmaI-excised erythromycin resistance cassette from pBSLerm was ligated into BsmI-digested and T4 DNA polymerase-filled p7502HI0147. The sequence of p7502HI0147ermF was determined in order to verify the correct position and orientation of the erythromycin gene. p7502HI0147ermF was digested with BstXI, and the 2,989-base-pair fragment that contained only the cloned H. influenzae DNA sequence and a small portion (17 base pairs) of the vector was isolated from an agarose gel and used to transform NTHi 2019. A 818-base-pair fragment of the erythromycin resistance gene was amplified from the genomic DNA of the putative mutants by using primers pBSLerm-up (5'-GGAGGAAAAAATAAAGAGGGTTATAATGAACGAG-3') and pBSLerm-down (5'-CACAAAAAATAGGTACACGAAAAACAAGTTAAGGG-3'), while no product was amplified from the wild-type NTHi 2019 genomic DNA. PCR amplification of the putative mutant genomic DNA with primers 147-up and 147-down (described above), amplified a 2,944-bp fragment, while a 1,762-bp product was amplified from the wild-type genomic DNA. The difference in the sizes (1,182 base pairs) is consistent with the size of the SmaI-excised erythromycin resistance gene. The mutants were verified by Southern blotting using a digoxigenin-labeled probe for detection of the erythromycin resistance gene and HI0147 (Roche Diagnostics Inc., Indianapolis, IN). This mutant strain was designated NTHi 2019siaT.
Colony blotting. NTHi 2019siaT mutants were grown on supplemented Difco brain heart infusion agar containing 5 μg/ml of erythromycin and 100 μM sialic acid. The wild-type NTHi 2019 was grown on S-BHI containing 100 μM sialic acid. Colony lifts were performed using nitrocellulose filters (Protran; 82-mm-diameter nitrocellulose disks; pore size, 0.45 μm; Schleicher & Schuell, Keene, NH) cut into quarters and placed on the bacterial plates in a region where individual colonies could be seen. After 1 min, the membranes were removed and dried overnight at room temperature (RT). The next day, the filters were blocked using two 60-min incubations in 20 mM Tris-500 mM NaCl (pH 7.45)-0.5% Tween 20 (TBST) with 1.0% bovine serum albumin (TBST-BSA). The filters pieces were rinsed 5 min in neuraminidase buffer (50 mM sodium acetate, 154 mM sodium chloride, 9 mM calcium chloride, 25 mg/ml human serum albumin [pH 5.6]). The quarter filters were cut in half, and one piece from each was incubated with light agitation overnight either in neuraminidase buffer or in neuraminidase buffer containing 0.05 U/ml neuraminidase (sialidase; Roche, Indianapolis, IN). The next day, the filter pieces were washed three times for 10 min in TBS. The filter pieces were washed once for 10 min in TBST-BSA, then incubated for 3 h at RT in monoclonal antibody 3F11 diluted 1:100 in TBST-BSA. At the end of the 3 h, the filters were washed 3 times for 15 min in TBST. The filters were then incubated for 1 h at RT with peroxidase labeled goat anti-mouse immunoglobulin M (IgM) (Kirkegaard and Perry, Gaithersburg, MD) diluted 1:10,000 in 0.5x TBST-BSA. After 1 h, the filter pieces were washed four times for 15 min in TBST. The filter pieces were incubated for 5 min in Super Signal West Pico (Pierce Chemical Co., Rockford, IL) and then exposed to film.
Reverse transcription-PCR of HI0148. Total bacterial RNA was isolated from wild-type and mutant strains of NTHi 2019 and 7502 using TRI reagent (Sigma Chemical Co., St. Louis, MO; T9424) and Sigma's protocols. Contaminating DNA was removed by digestion with DNase I (amplification grade; 18068-015; Invitrogen, Carlsbad, CA) according to Invitrogen's protocols. The RNA was further processed by using an RNeasy MinElute cleanup kit (QIAGEN, Valencia, Calif.) and QIAGEN's protocols. cDNA was created using SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA) and primers 5'-TTGAGCTATGAACCATCAACC-3' and 5'-CGCCAGTGACTAAATAATGACC-3' with Invitrogen's protocols. A 293-base-pair fragment from the latter third of the HI0148 ORF was amplified from that cDNA by PCR using a hot-start protocol (HotStarTaq; QIAGEN, Valencia, CA). The product was visualized by electrophoresis in a 1.3% agarose gel.
LOS preparation and neuraminidase treatment. Organisms were grown on S-BHI solid medium in the presence or absence of supplemental Neu5Ac (100 μg/ml) unless otherwise stated. The organisms from 10 heavily streaked plates were suspended in 25 ml of phosphate-buffered saline (PBS) and pelleted by centrifugation. They were washed once with PBS and once with deionized water, then extracted with 25 ml of phenol after both were equilibrated to 65°C. This mixture was cooled on ice for 1 h and separated by low-speed centrifugation. The top aqueous layer was removed and saved. The phenol layer was back-extracted once with water at 65°C, cooled, and centrifuged, and the second aqueous layer was added to the first. The residual phenol was removed from the aqueous layer and the LOS by precipitating the LOS twice using 0.3 M sodium acetate (final concentration) and 2 volumes of 100% ethanol. This was put into a –80°C freezer overnight and then centrifuged at 15,000 x g for 30 min. To remove any contaminating lipoproteins, the LOS pellets were resuspended in 8 ml of buffer A (0.06 M Tris base, 10 mM EDTA, 2.0% sodium dodecyl sulfate [SDS], pH 6.8) and incubated in a boiling water bath for 5 to 10 min. The samples were allowed to cool, and proteinase K (Sigma Chemical Co., St. Louis, MO) was added to a final concentration of 12.5 μg/ml. The samples were incubated at 37°C for 16 to 24 h. The LOS was precipitated as described above. The LOS was washed three times by precipitation, as above, with ethanol to remove any residual SDS. After the last precipitation, the LOS was resuspended in water and centrifuged at 100,000 x g for 75 min twice. The pellets were resuspended in water, frozen, and then lyophilized. The dry LOS was stored at RT. For SDS-polyacrylamide gel electrophoretic (PAGE) analysis, LOS was resuspended at 1 mg/ml in water, and 10 μg was digested with 5 milliunits of neuraminidase in neuraminidase buffer and incubated at 37°C for 2 h.
SDS-PAGE, silver staining, and Western blotting. SDS-PAGE gels were prepared as described by Lesse et al. (16). The gel was loaded with 3 to 5 μl from each LOS preparation (100 nanograms of LOS). Silver staining was performed by the method of Tsai and Frasch (27). Western blotting was performed by the method of Towbin et al. (26). The monoclonal antibody 3F11 recognizes a terminal N-acetyllactosamine structure and has been characterized previously (31). Detection of the antibody was performed using a peroxidase-labeled goat anti-mouse IgM secondary antibody (Kirkegaard and Perry Laboratories) and Super Signal West Pico chemiluminescent substrate (Pierce). LOS from Neisseria gonorrhoeae strain PID2 was used as a molecular weight standard (23).
Preparation of O-LOS and neuraminidase treatment. To make the LOS more amenable to mass spectrometric analysis, O-linked fatty acids were removed from the lipid A moiety as previously described (8). The highly purified LOS (0.1 mg) was incubated in anhydrous hydrazine (50 μl; Aldrich) at 37°C for 35 min with mixing every 10 min. Samples were cooled on ice prior to and after the addition of ice-cold acetone (250 μl; Aldrich), then transferred to –20°C for 2 h. The quenched reaction mixture was centrifuged (12,000 x g) for 45 min at 4°C. The supernatant was removed, and the pelleted O-deacylated LOS (O-LOS) was dissolved in MilliQ H2O (50 μl) and evaporated on a speed vacuum system (Savant). To remove salts and other low-molecular-weight contaminants, the O-LOS (20 to 30 μg) was suspended on a nitrocellulose membrane (type VS; pore size, 0.025 μm; Millipore Corp.) over water for approximately 1 h. The desalted O-LOS was removed from the membrane, concentrated with a speed vacuum system, and analyzed by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF-MS). For removal of Neu5Ac, the O-LOS (20 to 30 μg) was digested in 10 mM ammonium acetate, pH 6.0, containing immobilized neuraminidase from Clostridium perfringens (type VI; Sigma) for 20 h at 30°C with shaking. The enzyme was pelleted by centrifugation, and the supernatant (15 μl) was transferred to a nitrocellulose membrane for drop dialysis. The desialylated O-LOS was concentrated and analyzed by MALDI-TOF-MS.
MALDI-TOF-MS of O-LOS. Dowex 50 beads (100 to 200 mesh, NH4+ form; Bio-Rad, Hercules, Calif.) were added to a mixture containing equal volumes of dialyzed O-LOS (2 μg/μl) and a saturated solution of 2,5 dihydroxybenzoic acid in acetone (Aldrich). Samples were spotted onto a stainless steel MALDI target and analyzed on a Voyager DESTR-Plus TOF instrument (Applied Biosystems) with a N2 laser (337 nm) in negative-ion mode with linear optics (8). The delay time was 165 ns, and the grid voltage was 94% of the full acceleration voltage (20 kV). Spectra were acquired, averaged, and mass calibrated with an external calibrant consisting of an equimolar mixture of angiotensin I, adrenocorticotropin fragment 18-39, and adrenocorticotropin fragment 7-38 (Bachem, Torrance, CA).
Dose dependence of sialic acid incorporation into LOS. Wild-type NTHi 2019 and 2019nanA were grown in supplemented RPMI as described by Coleman et al. (3), but modified to be sialic acid free according to Greiner et al. (9), with the addition of increasing concentrations of sialic acid (0, 0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, and 1.0 mM). The LOS was O deacylated using anhydrous hydrazine and subjected to MALDI-TOF-MS. Acquired mass spectra were integrated, and the sum of the area of peaks representing all the O-LOS glycoform ions (sialylated and asialylated) was calculated. The sum of the areas of the sialylated glycoforms was calculated and expressed as a percentage of the total area of all glycoforms.
[3H]Neu5Ac uptake assay. NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA were grown in sialic acid-free supplemented RPMI (3, 9) to mid-log phase (A600, 0.4 to 0.6). The bacteria were pelleted by centrifugation at 9300 x g for 1 min at RT. The bacterial pellets were resuspended to an A600 of 2.0 in 1.5 ml fresh supplemented RPMI in 1.5-ml microcentrifuge tubes. A ligand mixture was made by adding 9 μCi (4.5 x 10–10 moles) of [3H]Neu5Ac (ART153; ARC, St. Louis, MO) to 9 μl of 2.5 mM unlabeled Neu5Ac. The reaction mixture was made by adding 4 μl of ligand mixture to 1.5 ml of bacterial suspension (final concentrations, 3.3 μM unlabeled Neu5Ac and 0.07 μM [3H]Neu5Ac). As quickly as possible, 100-μl samples were removed to Nuclepore membranes and aspirated through the membranes. The membranes were washed with 2 ml of PBS, pH 7.4. Aliquots of the reaction mixture were removed at the selected time points (10 s and 0.5, 1, 2, 3, 4, 5, 7.5, 10, 15, and 20 min), aspirated through the membrane, and washed as described above. At the end of the time points, the membranes were removed from the vacuum manifold and counted in scintillation fluid.
Bactericidal assay. NTHi strains 2019, 2019nanA, 2019siaT, and 2019nanAsiaT were grown to early-log phase (A600, 0.2) in supplemented BHI broth. A 0.5-ml aliquot of each was centrifuged for 1 min at 10,000 rpm in a microcentrifuge (Brinkmann-Eppendorf, New York, N.Y.) at RT. The pellet was resuspended in 1 ml of phosphate-buffered salt solution (PBSS) consisting of 10 mM K2HPO4, 10 mM KH2PO4, 136 mM NaCl, 5 mM KCl, 1 mM CaCl2, 0.3 mM MgCl2 · 6H2O, 1 mM MgSO4 · 7H2O, and 0.01% BSA, pH 7.0.
The bactericidal assay, modified from that reported by Andreoni et al. (1), was carried out in a 96-well plate in 200 μl (final volume). Pooled normal human serum (a 20-donor pool of serum from human volunteers who had no previous history of neisserial infections) was diluted to 10% in PBSS. A control containing pooled normal human serum that had been heat inactivated for 30 min at 56°C was included in each experiment. A 10-μl volume (1 x 106 organisms) of the resuspended bacteria was diluted into 190 μl of PBSS, and serial 1/10 dilutions were made in PBSS. A 20-μl volume of each dilution was spread on S-BHI with or without appropriate antibiotic selection and grown overnight at 37°C under 5% CO2. The colonies in these reaction mixtures were counted and used as the initial CFU. A 10-μl volume of the bacterial stock was incubated in the diluted serum for 30 min with shaking at 200 rpm in a 37°C incubator (Inova 4080; New Brunswick Scientific, Edison, NJ). Serial 1/10 dilutions of the reaction mixtures were diluted into PBSS and were spread on S-BHI with or without appropriate antibiotic selection. These were grown overnight at 37°C under 5% CO2, and emerging colonies were counted the next day. The resulting CFU value was that recorded after 30 min. Killing was assessed by comparing the number of CFU from the 30-min serum incubation with the number of the initial CFU. Results were expressed as the log10 change in CFU at 30 min compared to the initial CFU. Statistical analysis of the data from bactericidal assays was carried out using the paired t test and analysis of variance functions found in GraphPad Prism, version 4 (GraphPad, San Diego, CA).
Biofilm growth in a continuous-flow chamber. NTHi strains 2019 and 2019siaT were grown in a continuous-flow chamber identical to that described by Davies et al. (5). Strains were grown to mid-log phase in RPMI 1640 medium (Gibco BRL, Grand Island, N.Y.) supplemented with protoporphyrin IX (1 μg/ml), hypoxanthine (0.1 mg/ml), uracil (0.1 mg/ml), -NAD (10 μg/ml), sodium pyruvate (0.8 mM), and sialic acid (100 μM). Cultures were diluted to an optical density of 0.25 at 600 nm to inoculate chambers. Chambers were filled with the prepared inoculum and incubated at 37°C for 1 h to allow for adherence of the bacteria to the coverslip. Supplemented RPMI medium was diluted 1:10 with PBS and was pumped into chambers at a flow rate of 180 μl/min. Biofilm was allowed to grow for 2 days. Chambers were photographed and then stained with the LIVE/DEAD BacLight viability stain (Molecular Probes, Eugene, OR) by following the manufacturer's protocol. The biofilm was visualized with a Bio-Rad 1024 laser scanning confocal microscope at a magnification of x20. A 3-dimensional representation of the image was produced using the high-resolution rendering function of the Volocity program (http://www.bucher.ch/index2.html). Image J was used to display the vertical cross section of the biofilms (http://rsb.info.nih.gov/ij/).
Nucleotide sequence accession numbers. GenBank accession numbers are as follows: NTHi 2019 siaT (HI0147), DQ054471; NTHi 3198 HI0147 ORF, DQ054469; NTHi 7502 HI0147 ORF, DQ054470.
RESULTS
Effect of the HI1104 mutation on LOS sialylation. Based on sequence comparison analysis with the E. coli nanT gene sequence, it was assumed that HI1104 encoded a sialic acid transporter. The HI1104 ORF is predicted to encode a protein of 407 amino acids. HI1104 was deleted in the NTHi 2019 strain, and the LOS from the resultant strain was analyzed by SDS-PAGE and, after O deacylation, MALDI-TOF-MS. SDS-PAGE suggested that there was very little difference in sialylation between mutant and wild-type LOS profiles, even when the strains were grown on a medium supplemented with sialic acid (Fig. 1A, compare lanes 1 to 4 with lanes 5 to 8). Treatment of the LOS with neuraminidase showed identical band shifts in the mutant and the wild type, confirming that there are sialylated glycoforms present in the HI1104 mutant (Fig. 1A, compare lanes 2 and 4 with lanes 6 and 8, respectively).
To confirm the SDS-PAGE analysis, we O deacylated the LOS isolated from the wild-type and HI1104 mutant bacteria grown in the presence of sialic acid and subjected the samples to analysis by MALDI-TOF-MS. Previous studies have shown that NTHi 2019 produces a complex mixture of LOS glycoforms (7, 9, 20). The major component of this mixture has been extensively studied and is known to consist of a lactose moiety (Gal1,4-Glc1-) linked in a 1,4 linkage to HepI of the characteristic core structure of H. influenzae [HepIII1,2-HepII1,3-HepI1,5-Kdo(P)-lipid A] (19, 20, 24). The wild-type NTHi 2019 O-LOS (Fig. 1B and Table 2) gave a similar repertoire of glycoforms as that seen previously by MALDI-TOF-MS (9), with the major glycoform being the lactose-containing glycoform modified with 2 or 3 phosphoethanolamine (PEA) moieties (B2 and B3, respectively). Various larger glycoforms are present which differ from the B glycoforms by the addition of as many as four hexoses and a single N-acetylhexosamine. Additionally, many of the glycoforms can be decorated with up to two sialic acid moieties; such glycoforms disappear upon neuraminidase treatment of the O-LOS (data not shown). The MALDI-TOF-MS spectra of the HI1104 mutant (Fig. 1C; Table 2) appear to be very similar to the wild-type spectra, expressing both asialylated and, importantly, sialylated glycoforms. further demonstrating that HI1104 does not function as the sole sialic acid transporter in NTHi 2019. These data suggested that either multiple sialic acid transporters are present in NTHi or this gene does not encode the functional sialic acid transporter of NTHi.
Dose dependence of incorporation of sialic acid into LOS. To study the efficiency of sialic acid uptake in NTHi 2019, the wild type and a mutant lacking the neuraminyl lyase gene (nanA) were grown in defined sialic acid-free medium containing increasing amounts of sialic acid (0 to 1.0 mM). The LOS was then isolated from these bacteria, O deacylated with anhydrous hydrazine, and analyzed by MALDI-TOF-MS. The nanA mutant was included to create a condition where the sole fate of the sialic acid was incorporation into LOS. In the absence of sialic acid, the LOS profiles for the wild type and nanA mutant are almost identical (Fig. 2A). There are differences in the diversity and expression of the glycoforms observed in this experiment compared to the O-LOS in Fig. 1B. These differences are thought to occur due to whether the bacteria grow adhered to a surface (plates) or planktonically (liquid medium) (9). When the bacteria are grown in the presence of 1 mM sialic acid, the spectra of the O-LOS differ significantly (Fig. 2C and D). Although the wild type and the nanA mutant express the same glycoforms, the asialylated B glycoforms no longer predominate as the major LOS glycoform from the nanA mutant; rather, the disialylated B3 glycoform predominates (compare Fig. 2C and D). This confirms the observations of Vimr et al. (28) that the nanA mutant hypersialylates its LOS. When the ion peaks from the spectra are integrated and the sialylated glycoforms are expressed as a percentage of the total glycoforms (Fig. 2E), we see that the wild type is much less sensitive to increases in sialic acid concentration in the medium, suggesting that the nanA sialic acid degradation pathway is in balance with the CMP-sialic acid synthetase "activation" pathway, thus limiting the appearance of sialic acid in the LOS. The nanA mutant likely gives a more accurate indication of the efficiency of sialic acid transport, showing that LOS glycoforms containing sialic acid can be identified at sialic acid concentrations as low as 0.1 μM.
Cloning and mutagenesis of siaT. Figure 3A shows the genomic region surrounding siaT. The siaT ORF from the H. influenzae Rd genome database encodes a predicted protein of 633 amino acids. It was identified by homology with genes known to encode the transmembrane components of TRAP-type C4 dicarboxylate transporters from Vibrio cholerae (VC1777), Vibrio vulnificus (VVA1200), Haemophilus somnus (2336_DctQ), Pasturella multocida (PM1078), and Fusobacterium nucleatum (FN1473) (for sequence alignments, see Fig. 1S in the supplemental material). The siaT gene appears to be a fusion of the two membrane-spanning components of the TRAP-type transport system. The proximity of this gene encoding a potential transporter to the nan operon suggests that this gene may be involved in the uptake of environmental sialic acid. The predicted amino acid sequences of the siaT (HI0147) genes from NTHi 2019, 3198, and 7502 were aligned. The 7502 sequence differs from 2019 only in two amino acids (Ile13Met and Ser29Ala) (see Fig. 1S in the supplemental material). Plasmid p7502HI0147 was used to create the deletion mutant in all NTHi backgrounds, because it was the only plasmid that had convenient restriction enzyme sites near the center of the insert sequence. Transformants of 2019siaT were screened using a colony blot assay with monoclonal antibody 3F11, which is specific for an N-acetyllactosamine epitope (31) that is masked by the addition of sialic acid to the LOS. In the wild type, the 3F11 epitope is observed only upon treatment of the blotted colonies with neuraminidase; however, in the two transformants, 3F11 binding was observed both before and after neuraminidase treatment, suggesting that these mutants lack sialic acid and thus display N-acetyllactosamine as a terminal disaccharide on their LOS. The expression of HI0148, annotated as a hypothetical protein, 168 base pairs upstream of siaT was confirmed by reverse transcription-PCR (data not shown). siaT mutants were also created in NTHi 3198 and 7502 backgrounds.
Uptake of [3H]sialic acid by NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA. NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA were studied for their abilities to transport [3H]sialic acid. The 3H label of the sialic acid used in the uptake assay is located on C-9 and thus forms part of the pyruvate when the sialic acid is metabolized by the neuraminyl lyase (NanA), ultimately being lost as 3H2O. Thus, the uptake assay is measuring the incorporation of sialic acid into the LOS and does not take into account the sialic acid that enters the degradation pathway. Again, the neuraminyl lyase (nanA) mutants were included in this study to ensure that all of the transported sialic acid was incorporated into the LOS rather than into degradation pathways (28). Figure 3 shows the results of these studies. As can be seen, 2019siaT and 2019siaTnanA were unable to transport sialic acid and failed to accumulate an appreciable amount of the sialic acid over the 20-min time course of the experiment. During this time, NTHi 2019 acquired 8- to 9-fold more sialic acid than NTHi 2019siaT, while NTHi 2019nanA accumulated 80-fold more sialic acid than 2019siaTnanA. Additionally, the wild type rapidly achieved a steady state (<1 min), whereas the 2019nanA mutant continued to accumulate sialic acid, reaching saturation after 5 min. These studies indicate that the protein encoded by siaT is the sialic acid transporter in NTHi 2019. Uptake studies performed on siaT mutants of NTHi 3198 and NTHi 7502 gave similar results (data not shown).
Comparative analysis of LOS from NTHi 2019 and 2019siaT. LOS from wild-type NTHi 2019 and the siaT mutant were prepared from these strains grown on S-BHI agar with or without supplemental sialic acid. A portion of the LOS sample was treated with neuraminidase, and then pre- and post-neuraminidase treatment samples were resolved by SDS-PAGE (Fig. 4A). The wild-type NTHi 2019 gave a typical glycoform pattern for this strain, with a number of bands that became intensified upon growth of the bacteria with supplemental sialic acid (Fig. 4A, compare lanes 1 and 3). These intensified glycoforms disappeared upon treatment of the LOS with neuraminidase, and the acceptor glycoforms became intensified (Fig. 4A, compare lanes 1 and 3 with 2 and 4, respectively). In contrast, the glycoforms observed for the 2019siaT mutant remained the same regardless of whether or not supplemental sialic acid was added to the growth medium (Fig. 4A, compare lanes 5 and 7). Equally, there were no differences in the LOS profiles after neuraminidase treatment (Fig. 4A, compare lanes 5 and 7 with 6 and 8, respectively). It seems, upon comparison of the LOS of the mutant with that of the wild type, that the profile of the mutant is more similar to that of the neuraminidase-treated wild-type LOS. Taken together, these data suggest that in this mutant, sialic acid is not incorporated into the LOS. Western blot analysis using 3F11 was carried out on the wild-type and 2019siaT mutant LOS. The wild-type LOS grown on BHI (which contains trace amounts of sialic acid) without supplemental sialic acid was negative for 3F11 binding, suggesting that enough sialic acid could be incorporated into the LOS to mask the 3F11 epitope (Fig. 4B, lane 1). This was also the case when the wild-type bacteria were grown on BHI with supplemental sialic acid (Fig. 4B, lane 3). When treated with neuraminidase, wild-type 2019 grown under both conditions showed similar banding patterns, corresponding to LOS glycoforms containing terminal N-acetyllactosamine (Fig. 4B, lanes 2 and 4). For the 2019siaT mutant, bands were detected even when the bacteria were grown with supplemental sialic acid, suggesting an inability to mask this epitope (Fig. 4B, lanes 5 and 7). The number of bands detected after neuraminidase treatment remained the same, showing that no further epitopes were unmasked after this treatment (Fig. 4B, lanes 6 and 8). Interestingly, there were striking differences between the epitopes present in the wild type and the 2019siaT mutant after neuraminidase treatment, suggesting that there may be regulation of which glycoforms are expressed when sialic acid is not available to the bacteria (Fig. 4B, compare lanes 2 and 4 with lanes 6 and 8, respectively).
MALDI-TOF-MS of O-deacylated LOS. To further investigate the LOS phenotype, wild-type and siaT mutant LOS were O deacylated by treatment with anhydrous hydrazine and analyzed by MALDI-TOF-MS. MALDI-TOF-MS of NTHi 2019 (Fig. 5A and Table 2) gave a similar repertoire of glycoforms as that seen in the HI1104 study in Fig. 2A to D. The proportion of sialylated glycoforms present increased when the wild-type bacteria were grown in the presence of sialic acid (Fig. 5B). The MALDI-TOF-MS spectra of the O-LOS from 2019siaT showed a similar diversity in glycoforms; however, the mutant lacked the sialic acid-containing LOS glycoforms that could be seen in the wild type (Fig. 5C). These sialylated species were also completely absent from the O-LOS of 2019siaT grown on sialic acid-supplemented medium (Fig. 5D), indicating that in the absence of siaT, the bacteria were not capable of incorporating sialic acid into their LOS.
Bactericidal assay. It has been shown previously that there is a correlation between the incorporation of terminal sialic acid into LOS and protection of H. influenzae from complement-mediated killing by normal human serum. Since 2019siaT cannot acquire sialic acid for incorporation into LOS, it is likely that such a bacterium would have an increased susceptibility to killing by normal human serum. To investigate this, we carried out a bactericidal assay on the wild type, 2019siaT, 2019nanA, and 2019siaTnanA. In the absence of supplemental sialic acid in the medium, the wild type, 2019siaT, and 2019siaTnanA were all susceptible to killing by normal human serum (Fig. 6A). Interestingly, the 2019nanA mutant was resistant to serum killing, suggesting that the bacterium can acquire sufficient sialic acid from the BHI medium (which contains trace amounts of sialic acid) to afford protection from serum killing. The wild type and 2019nanA grown in a medium supplemented with 20 μM sialic acid were protected from serum killing by incorporation of sialic acid into their LOS (Fig. 6B). Conversely, 2019siaT and 2019siaTnanA, when grown in the presence of sialic acid, were still susceptible to serum killing (Fig. 6B). As a control, the experiment was repeated using heat-inactivated serum; as expected, both the wild type and mutants were capable of surviving this treatment (Fig. 6C and D). This finding clearly supports the MALDI-TOF-MS data and shows that in the absence of siaT, H. influenzae is incapable of incorporating sialic acid into its LOS.
Biofilm formation. Previous work has indicated that sialic acid is a significant component of the biofilm formed by NTHi 2019 (9). A continuous-flow cell assay was used to determine the ability of the 2019siaT mutant to form a biofilm. After 2 days of growth, both strains produced a visible biofilm (Fig. 7A and B). LIVE/DEAD staining revealed that a majority of cells in the wild-type biofilm were viable, while the 2019siaT biofilm had a higher proportion of dead cells. Additionally, the 2019siaT biofilm appeared to have a lower cell density than the wild-type biofilm, as judged by the signal intensity of the Syto-9 dye. While inactivation of siaT does not inhibit biofilm formation, the loss of sialic acid transport clearly has a profound effect on the viability of cells in the biofilm and the superstructure of the biofilm.
DISCUSSION
The surface of nontypeable Haemophilus influenzae is covered with LOS molecules. Incorporation of terminal sialic acid into the LOS enables NTHi to evade complement-dependent killing mechanisms, an important component of the host immune system (10, 11). The pathways that are involved in the incorporation of sialic acid into the LOS and/or its regulation are important targets for future drug development. Many of the genes involved in the metabolism of sialic acid in H. influenzae have been characterized, including those encoding the CMP-sialic acid synthetase (11), sialyltransferases (10, 12), and the neuraminyl lyase (28). As yet the sialic acid transporter of H. influenzae has not been characterized. Until recently it was thought that the sialic acid transporter of H. influenzae was encoded by the HI1104 ORF, a gene that has homology to the nanT gene of E. coli (30). The nanT gene product is a secondary transporter of the major facilitator superfamily that imports sialic acid in symport with a proton. Deletion of the HI1104 nanT homolog in NTHi 2019 had little effect on the ability of the bacteria to incorporate sialic acid into their LOS. H. influenzae lacks a sialic acid synthesis pathway, suggesting that either HI1104 is not a sialic acid transporter in NTHi 2019 or there is a second transporter that is capable of importing sialic acid into the bacteria.
To gain further insight into the efficiency of sialic acid uptake by NTHi 2019, the dose dependence of incorporation of sialic acid into the LOS was investigated. The process appears to be highly efficient in the nanA mutant background, creating a condition where all sialic acid is directed toward LOS sialylation; sialic acid can be detected in the LOS when the concentration of sialic acid in the medium is as low as 0.1 μM. More recently, it has been suggested in the literature that H. influenzae has a sialic acid transporter which is different from that of E. coli nanT (30). The genes HI0146 and HI0147 (siaT) have a close proximity in the H. influenzae genome to the nan operon, which encodes the genes responsible for the catabolism of sialic acid. By homology, these genes were predicted to encode an ESR and the transporter domains, respectively, of a transporter of the TRAP family (14, 21). Such transporters have been reported previously in bacteria and are often involved in the transport of C4 dicarboxylates such as succinate and fumarate using the electrochemical proton gradient as a driving force (6). To investigate the hypothesis that siaT acts as a sialic acid transporter in NTHi, the siaT gene was deleted in NTHi strains 2019, 3198, and 7502.
The sialic acid uptake assay demonstrates that the siaT gene product is likely the sialic acid transporter of NTHi. The 2019siaT and 2019siaTnanA mutants were both incapable of sialic acid uptake, indicating that the siaT gene product is required prior to activation of sialic acid by the CMP-sialic acid synthetase, thus implicating it as the transporter in NTHi 2019. The same assay repeated on siaT mutants in NTHi strains 3198 and 7502 gave similar results (data not shown). The assay also demonstrates some interesting aspects of the uptake and incorporation of sialic acid into the LOS of H. influenzae. In the wild type, the amount of sialic acid detected rapidly reaches a "steady-state" level, in contrast to the 2019nanA mutant, which continues to accumulate sialic acid until reaching saturation. Indeed, it has been shown previously that H. influenzae nanA mutants hypersialylate their LOS (28). This would suggest that the point of regulation of sialic acid levels within the bacteria occurs not through regulation of the transporter but through regulation of the downstream gene products in the sialic acid pathway such as the neuraminyl lyase and the CMP-N-sialic acid synthetase. The concept that the sialic acid transporter is expressed at a constitutive level would make some sense, because sialic acid, as well as being an important molecule in the evasion of the host immune response, is a valuable carbon and nitrogen source.
Deletion of siaT, as suggested by the sialic acid uptake data, results in bacteria that are incapable of incorporating sialic acid into their LOS, thus suggesting that this gene is indeed involved in the uptake of sialic acid in NTHi. This conclusion is supported by the SDS-PAGE and MALDI-TOF-MS data, all of which conclusively indicate the lack of sialic acid-containing LOS glycoforms on the 2019siaT mutant. As a consequence, the siaT mutant is severely compromised in its ability to evade the host immune response, as evidenced by the fact that the mutant is susceptible to complement-mediated killing when exposed to normal human sera. The Western blot analysis suggested that the 2019siaT mutant was not only compromised in its ability to make sialylated LOS glycoforms but also expressed fewer glycoforms containing terminal N-acetyllactosamine. This suggests that in the absence of sialic acid, the bacteria express different LOS glycoforms, although the mechanism of this regulation remains to be investigated.
A point of secondary interest is that the 2019nanA mutant, when grown on BHI medium (which contains trace amounts of sialic acid), can acquire enough sialic acid from the medium to produce LOS that is sufficiently sialylated to protect the bacterium from complement-mediated lysis, thus indicating that the deletion of nanA leads to an increase in "flux" of sialic acid into the LOS incorporation pathway.
Finally, in the absence of Neu5Ac transport, NTHi 2019siaT, although capable of forming a biofilm, produces a biofilm with altered morphology, lower cell density, and a higher proportion of dead cells than wild-type NTHi 2019 biofilm. We are uncertain why bacterial viability is reduced when Neu5Ac is limiting. A similar phenotype was seen with 2019siaA during middle ear infection in a chinchilla model of otitis media (13).
This is the first TRAP transporter to be characterized in H. influenzae and also the first TRAP transporter known to transport sialic acid. The best-described and -characterized TRAP transporter is the DctPQM C4 dicarboxylate transporter of Rhodobacter capsulatus, though various homologs have been identified in archaea and gram-negative bacteria. DctP is the periplasmic extracellular solute receptor, while DctQ and DctM represent the integral membrane proteins with 4 and 12 predicted membrane-spanning regions, respectively (6). In H. influenzae these two membrane proteins are encoded by a single gene containing a total of 16 predicted membrane-spanning regions (21). The ESR protein is thought to increase the uptake affinity of the transporter by binding sialic acid and delivering it to the transporter. Such a high-affinity sialic acid uptake system may be important for the bacteria in their normal physiological environment. The novelty of this transporter may make it an important drug target given the dependence of H. influenzae on sialic acid for immune evasion.
The diversity that exists among sialic acid transporters in gram-negative bacteria, which so far include members of the major facilitator family and TRAP transporters, is noteworthy. This diversity may be driven, in part, by the particular environments to which bacteria have adapted, the challenges these environments present for their growth and survival, and the abundance of sialic acid that is present in these environments. For example, although the nasopharynx is rich in secreted sialylated mucins and possibly other types of sialoglycoconjugates, free sialic acid is much more limited, and the mechanism by which H. influenzae may exploit these more abundant bound sources of sialic acid remains to be elucidated.
In conclusion, deletion of the siaT gene of H. influenzae leads to bacteria that are not capable of sialic acid uptake; thus, the bacteria cannot sialylate their LOS, making them vulnerable to complement-mediated killing. The siaT gene product therefore appears to be the sole sialic acid transporter of the NTHi strains investigated. The siaT gene product is a transporter of the TRAP transporter family, making this the first sialic acid transporter of this type to be investigated.
ACKNOWLEDGMENTS
This work was supported by grants from the National Institutes of Health (AI24616 to M.A.A. and B.W.G.; AI30040 to M.A.A.).
Supplemental material for this article may be found at http://iai.asm.org/.
REFERENCES
1. Andreoni, J., H. Kayhty, and P. Densen. 1993. Vaccination and the role of capsular polysaccharide antibody in prevention of recurrent meningococcal disease in late complement component-deficient individuals. J. Infect. Dis. 168:227-231.
2. Campagnari, A. A., M. R. Gupta, K. C. Dudas, T. F. Murphy, and M. A. Apicella. 1987. Antigenic diversity of lipooligosaccharides of nontypeable Haemophilus influenzae. Infect. Immun. 55:882-887.
3. Coleman, H. N., D. A. Daines, J. Jarisch, and A. L. Smith. 2003. Chemically defined media for growth of Haemophilus influenzae strains. J. Clin. Microbiol. 41:4408-4410.
4. Davidson, A. L., and J. Chen. 2004. ATP-binding cassette transporters in bacteria. Annu. Rev. Biochem. 73:241-268.
5. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298.
6. Forward, J. A., M. C. Behrendt, N. R. Wyborn, R. Cross, and D. J. Kelly. 1997. TRAP transporters: a new family of periplasmic solute transport systems encoded by the dctPQM genes of Rhodobacter capsulatus and by homologs in diverse gram-negative bacteria. J. Bacteriol. 179:5482-5493.
7. Gaucher, S. P., M. T. Cancilla, N. J. Phillips, B. W. Gibson, and J. A. Leary. 2000. Mass spectral characterization of lipooligosaccharides from Haemophilus influenzae 2019. Biochemistry 39:12406-12414.
8. Gibson, B. W., J. J. Engstrom, C. M. John, W. Hines, and A. M. Falick. 1997. Characterization of bacterial lipooligosaccharides by delayed extraction matrix-assisted laser desorption time-of-flight mass spectrometry. J. Am. Soc. Mass Spectrom. 8:645-658.
9. Greiner, L. L., H. Watanabe, N. J. Phillips, J. Shao, A. Morgan, A. Zaleski, B. W. Gibson, and M. A. Apicella. 2004. Nontypeable Haemophilus influenzae strain 2019 produces a biofilm containing N-acetylneuraminic acid that may mimic sialylated O-linked glycans. Infect. Immun. 72:4249-4260.
10. Hood, D. W., A. D. Cox, M. Gilbert, K. Makepeace, S. Walsh, M. E. Deadman, A. Cody, A. Martin, M. Mansson, E. K. Schweda, J. R. Brisson, J. C. Richards, E. R. Moxon, and W. W. Wakarchuk. 2001. Identification of a lipopolysaccharide -2,3-sialyltransferase from Haemophilus influenzae. Mol. Microbiol. 39:341-350.
11. Hood, D. W., K. Makepeace, M. E. Deadman, R. F. Rest, P. Thibault, A. Martin, J. C. Richards, and E. R. Moxon. 1999. Sialic acid in the lipopolysaccharide of Haemophilus influenzae: strain distribution, influence on serum resistance and structural characterization. Mol. Microbiol. 33:679-692.
12. Jones, P. A., N. M. Samuels, N. J. Phillips, R. S. Munson, Jr., J. A. Bozue, J. A. Arseneau, W. A. Nichols, A. Zaleski, B. W. Gibson, and M. A. Apicella. 2002. Haemophilus influenzae type b strain A2 has multiple sialyltransferases involved in lipooligosaccharide sialylation. J. Biol. Chem. 277:14598-14611.
13. Jurcisek, J., L. Greiner, H. Watanabe, A. Zaleski, M. A. Apicella, and L. O. Bakaletz. 2005. Role of sialic acid and complex carbohydrate biosynthesis in biofilm formation by nontypeable Haemophilus influenzae in the chinchilla middle ear. Infect. Immun. 73:3210-3218.
14. Kelly, D. J., and G. H. Thomas. 2001. The tripartite ATP-independent periplasmic (TRAP) transporters of bacteria and archaea. FEMS Microbiol. Rev. 25:405-424.
15. Kolker, E., K. S. Makarova, S. Shabalina, A. F. Picone, S. Purvine, T. Holzman, T. Cherny, D. Armbruster, R. S. Munson, Jr., G. Kolesov, D. Frishman, and M. Y. Galperin. 2004. Identification and functional analysis of ‘hypothetical’ genes expressed in Haemophilus influenzae. Nucleic Acids Res. 32:2353-2361.
16. Lesse, A. J., A. A. Campagnari, W. E. Bittner, and M. A. Apicella. 1990. Increased resolution of lipopolysaccharides and lipooligosaccharides utilizing tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis. J. Immunol. Methods 126:109-117.
17. Mandrell, R. E., R. McLaughlin, Y. Aba Kwaik, A. Lesse, R. Yamasaki, B. Gibson, S. M. Spinola, and M. A. Apicella. 1992. Lipooligosaccharides (LOS) of some Haemophilus species mimic human glycosphingolipids, and some LOS are sialylated. Infect. Immun. 60:1322-1328.
18. Mansson, M., S. H. Bauer, D. W. Hood, J. C. Richards, E. R. Moxon, and E. K. Schweda. 2001. A new structural type for Haemophilus influenzae lipopolysaccharide. Structural analysis of the lipopolysaccharide from nontypeable Haemophilus influenzae strain 486. Eur. J. Biochem. 268:2148-2159.
19. Masoud, H., E. R. Moxon, A. Martin, D. Krajcarski, and J. C. Richards. 1997. Structure of the variable and conserved lipopolysaccharide oligosaccharide epitopes expressed by Haemophilus influenzae serotype b strain Eagan. Biochemistry 36:2091-2103.
20. Phillips, N. J., M. A. Apicella, J. M. Griffiss, and B. W. Gibson. 1992. Structural characterization of the cell surface lipooligosaccharides from a nontypable strain of Haemophilus influenzae. Biochemistry 31:4515-4526.
21. Rabus, R., D. L. Jack, D. J. Kelly, and M. H. Saier, Jr. 1999. TRAP transporters: an ancient family of extracytoplasmic solute-receptor-dependent secondary active transporters. Microbiology 145:3431-3445.
22. Rao, V. K., G. P. Krasan, D. R. Hendrixson, S. Dawid, and J. W. St Geme III. 1999. Molecular determinants of the pathogenesis of disease due to non-typable Haemophilus influenzae. FEMS Microbiol. Rev. 23:99-129.
23. Schneider, H., J. M. Griffiss, J. W. Boslego, P. J. Hitchcock, K. M. Zahos, and M. A. Apicella. 1991. Expression of paragloboside-like lipooligosaccharides may be a necessary component of gonococcal pathogenesis in men. J. Exp. Med. 174:1601-1605.
24. Schweda, E. K., O. E. Hegedus, S. Borrelli, A. A. Lindberg, J. N. Weiser, D. J. Maskell, and E. R. Moxon. 1993. Structural studies of the saccharide part of the cell envelope lipopolysaccharide from Haemophilus influenzae strain AH1-3 (lic3+). Carbohydr. Res. 246:319-330.
25. Teele, D. W., J. O. Klein, B. Rosner, L. Bratton, G. R. Fisch, O. R. Mathieu, P. J. Porter, S. G. Starobin, L. D. Tarlin, and R. P. Younes. 1983. Middle ear disease and the practice of pediatrics. Burden during the first five years of life. JAMA 249:1026-1029.
26. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.
27. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 119:115-119.
28. Vimr, E., C. Lichtensteiger, and S. Steenbergen. 2000. Sialic acid metabolism's dual function in Haemophilus influenzae. Mol. Microbiol. 36:1113-1123.
29. Vimr, E. R., K. A. Kalivoda, E. L. Deszo, and S. M. Steenbergen. 2004. Diversity of microbial sialic acid metabolism. Microbiol. Mol. Biol. Rev. 68:132-153.
30. Vimr, E. R., and F. A. Troy. 1985. Identification of an inducible catabolic system for sialic acids (nan) in Escherichia coli. J. Bacteriol. 164:845-853.
31. Yamasaki, R., W. Nasholds, H. Schneider, and M. A. Apicella. 1991. Epitope expression and partial structural characterization of F62 lipooligosaccharide (LOS) of Neisseria gonorrhoeae: IgM monoclonal antibodies (3F11 and 1-1-M) recognize non-reducing termini of the LOS components. Mol. Immunol. 28:1233-1242.(Simon Allen, Anthony Zale)
Department of Microbiology, University of Iowa, Iowa City, Iowa 52242
Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94143-0446
ABSTRACT
Nontypeable Haemophilus influenzae is an opportunistic pathogen and a common cause of otitis media in children and of chronic bronchitis and pneumonia in patients with chronic obstructive pulmonary disease. The lipooligosaccharides, a major component of the outer membrane of H. influenzae, play an important role in microbial virulence and pathogenicity. N-Acetylneuraminic acid (sialic acid) can be incorporated into the lipooligosaccharides as a terminal nonreducing sugar. Although much of the pathway of sialic acid incorporation into lipooligosaccharides is understood, the transporter responsible for N-acetylneuraminic acid uptake in H. influenzae has yet to be characterized. In this paper we demonstrate that this transporter is a novel sugar transporter of the tripartite ATP-independent periplasmic transporter family. In the absence of this transporter, H. influenzae cannot incorporate sialic acid into its lipooligosaccharides, making the organism unable to survive when exposed to human serum and causing reduced viability in biofilm growth.
INTRODUCTION
Haemophilus influenzae is exclusively adapted to infect or colonize humans. Strains can be encapsulated or nonencapsulated (nontypeable). Nontypeable Haemophilus influenzae (NTHi) is a frequent colonizer of the nasopharynx. When the airway is compromised, NTHi can cause local infections such as otitis media in young children (24 million physician visits per year in the United States [25]) and chronic bronchitis and pneumonia in patients with chronic obstructive pulmonary disease.
Lipooligosaccharides (LOS) are a major component of the NTHi outer membrane and have been shown to play a role in microbial virulence and pathogenicity (22). LOS contains carbohydrate epitopes which mimic human glycosphingolipids, allowing the bacteria to avoid the host immune response (17). LOS present on the surface of NTHi is a heterogeneous mixture of glycoforms, the most abundant of which has been extensively studied and is known to consist of a lactose moiety (Gal1-4Glc) attached to the first heptose (HepI) of a conserved core structure [HepIII1,2-HepII1,3-HepI1,5-Kdo(P)-lipid A] (19, 20, 24). Importantly, NTHi is also capable of incorporating the acidic sugar N-acetylneuraminic acid (Neu5Ac, or sialic acid) as terminal nonreducing units into its LOS, giving the bacterium protection from complement-mediated killing by normal human serum (10, 11). The acceptors for sialic acid are lactose (18), N-acetyllactosamine, and possibly N-acetylgalactosamine, although the precise structures of most of these sialylated LOS species have not been conclusively identified (12).
The sialic acid is incorporated into the LOS before it reaches the cell surface by one of three sialyltransferases, SiaA, Lic3a, or LsgB (10, 12). The donor for this transfer is CMP-sialic acid, which is synthesized from sialic acid and CTP by the CMP-sialic acid synthetase (SiaB) (11). The fate of sialic acid in NTHi is not solely incorporation into the LOS; sialic acid can also be utilized as a carbon and nitrogen source via its breakdown to N-acetylmannosamine and pyruvate by the neuraminyl lyase (NanA) (28).
NTHi is incapable of synthesizing sialic acid and thus requires an exogenous source of sialic acid for incorporation to occur. In Escherichia coli, sialic acid is imported via symport with a proton through a specific transporter (NanT) of the major facilitator superfamily (30). A gene (HI1104) was identified in the H. influenzae genome that has high homology to the E. coli sialic acid transporter. HI1104 was deleted in this study and shown to have no effect on sialic acid uptake in the H. influenzae strain studied here. Recent publications have suggested that sialic acid transport in H. influenzae is mediated via a novel class of transporter, a tripartite ATP-independent periplasmic (TRAP) transporter (15, 29). TRAP transporters consist of three components: an extracellular solute receptor (ESR) and two distinct integral membrane components of unequal size which are sometimes fused (14). These transporters differ from the better-characterized ABC-protein transporter family (4) in that they do not possess an ATP-binding cassette protein and are not driven by ATP hydrolysis but rather by an electrochemical ion gradient (14).
The gene HI0147 (siaT) was previously identified by Rabus et al. (21) as the fused transmembrane domains of a TRAP transporter and named Y147; it was recently suggested to be part of a sialic acid transporter (15, 29). In this paper we provide evidence that the siaT gene product is indeed a component of the sialic acid TRAP transporter in the NTHi strain 2019. Deletion of the gene encoding this protein has a marked effect on the incorporation of sialic acid into the LOS and on the survival of the organism when exposed to human serum.
MATERIALS AND METHODS
Bacterial growth. Strains and plasmids used in this study are listed in Table 1. Escherichia coli was grown at 37°C in Luria-Bertani medium with or without agar (1.5%) and supplemented with antibiotics as needed. Wild-type H. influenzae was grown on supplemented brain heart infusion (BHI) agar (Difco Laboratories, Detroit, MI) supplemented with 10 μg/ml hemin and 10 μg/ml NAD at 37°C. Erythromycin-resistant H. influenzae was selected on supplemented BHI agar with 5 μg of erythromycin/ml. Selection was carried out without CO2.
Cloning and mutagenesis of HI1104. PCR was used to amplify a 2,146-bp fragment from NTHi 2019 genomic DNA containing the HI1104 open reading frame (ORF) by using oligonucleotide primers (5'-TCCCCCCGGGTCATGGAAAGATACGGATGCAAAG-3' and 5'-TCCCCCCGGGTCAAAAGGCGACAAAGAGGGTGG-3') with restriction sites for SmaI (underlined). This fragment was digested with SmaI and cloned into the SmaI site in pACYC177 (New England Biolabs, Beverly, MA). The sequence of the fragment was confirmed by sequencing and comparison with the H. influenzae Rd Kw-20 genome. This construct was named p20191104. The SmaI fragment of pBSLerm containing the erythromycin resistance gene (mlsR) was inserted into StyI/BsaBI-digested and blunt end-filled p20191104 by eliminating 759 of 1,220 bp of the 2019 HI1104 ORF. This plasmid was designated p20191104erm. NTHi 2019 was transformed with the 2,882-bp SmaI fragment of p20191104erm that was isolated away from the plasmid backbone by electrophoresis in an agarose gel. The NTHi mutant was confirmed by PCR and Southern blot analysis.
Cloning and mutation of NTHi 2019siaT. A 1,762-base-pair DNA fragment was amplified from genomic DNA of NTHi strains 2019, 3198, and 7502 by PCR with primers 147-up (5'-TTTCCTACACGAGCAACAAC-3') and 147-down (5'-CTACATTCCCTTATTCTTCATCAAAC-3'). This fragment was cloned by ligation into the pCR2.1TOPO vector, and transformation of DH5 host bacteria was carried out using the manufacturer's protocols (Invitrogen, Carlsbad, CA). The sequences of the TA inserts were determined. These fragments corresponded to bases 16 to 1778 of the 1,902-base-pair HI0147 open reading frame from the complete genome of H. influenzae Rd KW-20 (www.tigr.org). These plasmids were named p2019HI0147, p3198HI0147, and p7502HI0147.
Only p7502HI0147 had a convenient restriction enzyme site near the center of the insert sequence. The SmaI-excised erythromycin resistance cassette from pBSLerm was ligated into BsmI-digested and T4 DNA polymerase-filled p7502HI0147. The sequence of p7502HI0147ermF was determined in order to verify the correct position and orientation of the erythromycin gene. p7502HI0147ermF was digested with BstXI, and the 2,989-base-pair fragment that contained only the cloned H. influenzae DNA sequence and a small portion (17 base pairs) of the vector was isolated from an agarose gel and used to transform NTHi 2019. A 818-base-pair fragment of the erythromycin resistance gene was amplified from the genomic DNA of the putative mutants by using primers pBSLerm-up (5'-GGAGGAAAAAATAAAGAGGGTTATAATGAACGAG-3') and pBSLerm-down (5'-CACAAAAAATAGGTACACGAAAAACAAGTTAAGGG-3'), while no product was amplified from the wild-type NTHi 2019 genomic DNA. PCR amplification of the putative mutant genomic DNA with primers 147-up and 147-down (described above), amplified a 2,944-bp fragment, while a 1,762-bp product was amplified from the wild-type genomic DNA. The difference in the sizes (1,182 base pairs) is consistent with the size of the SmaI-excised erythromycin resistance gene. The mutants were verified by Southern blotting using a digoxigenin-labeled probe for detection of the erythromycin resistance gene and HI0147 (Roche Diagnostics Inc., Indianapolis, IN). This mutant strain was designated NTHi 2019siaT.
Colony blotting. NTHi 2019siaT mutants were grown on supplemented Difco brain heart infusion agar containing 5 μg/ml of erythromycin and 100 μM sialic acid. The wild-type NTHi 2019 was grown on S-BHI containing 100 μM sialic acid. Colony lifts were performed using nitrocellulose filters (Protran; 82-mm-diameter nitrocellulose disks; pore size, 0.45 μm; Schleicher & Schuell, Keene, NH) cut into quarters and placed on the bacterial plates in a region where individual colonies could be seen. After 1 min, the membranes were removed and dried overnight at room temperature (RT). The next day, the filters were blocked using two 60-min incubations in 20 mM Tris-500 mM NaCl (pH 7.45)-0.5% Tween 20 (TBST) with 1.0% bovine serum albumin (TBST-BSA). The filters pieces were rinsed 5 min in neuraminidase buffer (50 mM sodium acetate, 154 mM sodium chloride, 9 mM calcium chloride, 25 mg/ml human serum albumin [pH 5.6]). The quarter filters were cut in half, and one piece from each was incubated with light agitation overnight either in neuraminidase buffer or in neuraminidase buffer containing 0.05 U/ml neuraminidase (sialidase; Roche, Indianapolis, IN). The next day, the filter pieces were washed three times for 10 min in TBS. The filter pieces were washed once for 10 min in TBST-BSA, then incubated for 3 h at RT in monoclonal antibody 3F11 diluted 1:100 in TBST-BSA. At the end of the 3 h, the filters were washed 3 times for 15 min in TBST. The filters were then incubated for 1 h at RT with peroxidase labeled goat anti-mouse immunoglobulin M (IgM) (Kirkegaard and Perry, Gaithersburg, MD) diluted 1:10,000 in 0.5x TBST-BSA. After 1 h, the filter pieces were washed four times for 15 min in TBST. The filter pieces were incubated for 5 min in Super Signal West Pico (Pierce Chemical Co., Rockford, IL) and then exposed to film.
Reverse transcription-PCR of HI0148. Total bacterial RNA was isolated from wild-type and mutant strains of NTHi 2019 and 7502 using TRI reagent (Sigma Chemical Co., St. Louis, MO; T9424) and Sigma's protocols. Contaminating DNA was removed by digestion with DNase I (amplification grade; 18068-015; Invitrogen, Carlsbad, CA) according to Invitrogen's protocols. The RNA was further processed by using an RNeasy MinElute cleanup kit (QIAGEN, Valencia, Calif.) and QIAGEN's protocols. cDNA was created using SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA) and primers 5'-TTGAGCTATGAACCATCAACC-3' and 5'-CGCCAGTGACTAAATAATGACC-3' with Invitrogen's protocols. A 293-base-pair fragment from the latter third of the HI0148 ORF was amplified from that cDNA by PCR using a hot-start protocol (HotStarTaq; QIAGEN, Valencia, CA). The product was visualized by electrophoresis in a 1.3% agarose gel.
LOS preparation and neuraminidase treatment. Organisms were grown on S-BHI solid medium in the presence or absence of supplemental Neu5Ac (100 μg/ml) unless otherwise stated. The organisms from 10 heavily streaked plates were suspended in 25 ml of phosphate-buffered saline (PBS) and pelleted by centrifugation. They were washed once with PBS and once with deionized water, then extracted with 25 ml of phenol after both were equilibrated to 65°C. This mixture was cooled on ice for 1 h and separated by low-speed centrifugation. The top aqueous layer was removed and saved. The phenol layer was back-extracted once with water at 65°C, cooled, and centrifuged, and the second aqueous layer was added to the first. The residual phenol was removed from the aqueous layer and the LOS by precipitating the LOS twice using 0.3 M sodium acetate (final concentration) and 2 volumes of 100% ethanol. This was put into a –80°C freezer overnight and then centrifuged at 15,000 x g for 30 min. To remove any contaminating lipoproteins, the LOS pellets were resuspended in 8 ml of buffer A (0.06 M Tris base, 10 mM EDTA, 2.0% sodium dodecyl sulfate [SDS], pH 6.8) and incubated in a boiling water bath for 5 to 10 min. The samples were allowed to cool, and proteinase K (Sigma Chemical Co., St. Louis, MO) was added to a final concentration of 12.5 μg/ml. The samples were incubated at 37°C for 16 to 24 h. The LOS was precipitated as described above. The LOS was washed three times by precipitation, as above, with ethanol to remove any residual SDS. After the last precipitation, the LOS was resuspended in water and centrifuged at 100,000 x g for 75 min twice. The pellets were resuspended in water, frozen, and then lyophilized. The dry LOS was stored at RT. For SDS-polyacrylamide gel electrophoretic (PAGE) analysis, LOS was resuspended at 1 mg/ml in water, and 10 μg was digested with 5 milliunits of neuraminidase in neuraminidase buffer and incubated at 37°C for 2 h.
SDS-PAGE, silver staining, and Western blotting. SDS-PAGE gels were prepared as described by Lesse et al. (16). The gel was loaded with 3 to 5 μl from each LOS preparation (100 nanograms of LOS). Silver staining was performed by the method of Tsai and Frasch (27). Western blotting was performed by the method of Towbin et al. (26). The monoclonal antibody 3F11 recognizes a terminal N-acetyllactosamine structure and has been characterized previously (31). Detection of the antibody was performed using a peroxidase-labeled goat anti-mouse IgM secondary antibody (Kirkegaard and Perry Laboratories) and Super Signal West Pico chemiluminescent substrate (Pierce). LOS from Neisseria gonorrhoeae strain PID2 was used as a molecular weight standard (23).
Preparation of O-LOS and neuraminidase treatment. To make the LOS more amenable to mass spectrometric analysis, O-linked fatty acids were removed from the lipid A moiety as previously described (8). The highly purified LOS (0.1 mg) was incubated in anhydrous hydrazine (50 μl; Aldrich) at 37°C for 35 min with mixing every 10 min. Samples were cooled on ice prior to and after the addition of ice-cold acetone (250 μl; Aldrich), then transferred to –20°C for 2 h. The quenched reaction mixture was centrifuged (12,000 x g) for 45 min at 4°C. The supernatant was removed, and the pelleted O-deacylated LOS (O-LOS) was dissolved in MilliQ H2O (50 μl) and evaporated on a speed vacuum system (Savant). To remove salts and other low-molecular-weight contaminants, the O-LOS (20 to 30 μg) was suspended on a nitrocellulose membrane (type VS; pore size, 0.025 μm; Millipore Corp.) over water for approximately 1 h. The desalted O-LOS was removed from the membrane, concentrated with a speed vacuum system, and analyzed by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF-MS). For removal of Neu5Ac, the O-LOS (20 to 30 μg) was digested in 10 mM ammonium acetate, pH 6.0, containing immobilized neuraminidase from Clostridium perfringens (type VI; Sigma) for 20 h at 30°C with shaking. The enzyme was pelleted by centrifugation, and the supernatant (15 μl) was transferred to a nitrocellulose membrane for drop dialysis. The desialylated O-LOS was concentrated and analyzed by MALDI-TOF-MS.
MALDI-TOF-MS of O-LOS. Dowex 50 beads (100 to 200 mesh, NH4+ form; Bio-Rad, Hercules, Calif.) were added to a mixture containing equal volumes of dialyzed O-LOS (2 μg/μl) and a saturated solution of 2,5 dihydroxybenzoic acid in acetone (Aldrich). Samples were spotted onto a stainless steel MALDI target and analyzed on a Voyager DESTR-Plus TOF instrument (Applied Biosystems) with a N2 laser (337 nm) in negative-ion mode with linear optics (8). The delay time was 165 ns, and the grid voltage was 94% of the full acceleration voltage (20 kV). Spectra were acquired, averaged, and mass calibrated with an external calibrant consisting of an equimolar mixture of angiotensin I, adrenocorticotropin fragment 18-39, and adrenocorticotropin fragment 7-38 (Bachem, Torrance, CA).
Dose dependence of sialic acid incorporation into LOS. Wild-type NTHi 2019 and 2019nanA were grown in supplemented RPMI as described by Coleman et al. (3), but modified to be sialic acid free according to Greiner et al. (9), with the addition of increasing concentrations of sialic acid (0, 0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, and 1.0 mM). The LOS was O deacylated using anhydrous hydrazine and subjected to MALDI-TOF-MS. Acquired mass spectra were integrated, and the sum of the area of peaks representing all the O-LOS glycoform ions (sialylated and asialylated) was calculated. The sum of the areas of the sialylated glycoforms was calculated and expressed as a percentage of the total area of all glycoforms.
[3H]Neu5Ac uptake assay. NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA were grown in sialic acid-free supplemented RPMI (3, 9) to mid-log phase (A600, 0.4 to 0.6). The bacteria were pelleted by centrifugation at 9300 x g for 1 min at RT. The bacterial pellets were resuspended to an A600 of 2.0 in 1.5 ml fresh supplemented RPMI in 1.5-ml microcentrifuge tubes. A ligand mixture was made by adding 9 μCi (4.5 x 10–10 moles) of [3H]Neu5Ac (ART153; ARC, St. Louis, MO) to 9 μl of 2.5 mM unlabeled Neu5Ac. The reaction mixture was made by adding 4 μl of ligand mixture to 1.5 ml of bacterial suspension (final concentrations, 3.3 μM unlabeled Neu5Ac and 0.07 μM [3H]Neu5Ac). As quickly as possible, 100-μl samples were removed to Nuclepore membranes and aspirated through the membranes. The membranes were washed with 2 ml of PBS, pH 7.4. Aliquots of the reaction mixture were removed at the selected time points (10 s and 0.5, 1, 2, 3, 4, 5, 7.5, 10, 15, and 20 min), aspirated through the membrane, and washed as described above. At the end of the time points, the membranes were removed from the vacuum manifold and counted in scintillation fluid.
Bactericidal assay. NTHi strains 2019, 2019nanA, 2019siaT, and 2019nanAsiaT were grown to early-log phase (A600, 0.2) in supplemented BHI broth. A 0.5-ml aliquot of each was centrifuged for 1 min at 10,000 rpm in a microcentrifuge (Brinkmann-Eppendorf, New York, N.Y.) at RT. The pellet was resuspended in 1 ml of phosphate-buffered salt solution (PBSS) consisting of 10 mM K2HPO4, 10 mM KH2PO4, 136 mM NaCl, 5 mM KCl, 1 mM CaCl2, 0.3 mM MgCl2 · 6H2O, 1 mM MgSO4 · 7H2O, and 0.01% BSA, pH 7.0.
The bactericidal assay, modified from that reported by Andreoni et al. (1), was carried out in a 96-well plate in 200 μl (final volume). Pooled normal human serum (a 20-donor pool of serum from human volunteers who had no previous history of neisserial infections) was diluted to 10% in PBSS. A control containing pooled normal human serum that had been heat inactivated for 30 min at 56°C was included in each experiment. A 10-μl volume (1 x 106 organisms) of the resuspended bacteria was diluted into 190 μl of PBSS, and serial 1/10 dilutions were made in PBSS. A 20-μl volume of each dilution was spread on S-BHI with or without appropriate antibiotic selection and grown overnight at 37°C under 5% CO2. The colonies in these reaction mixtures were counted and used as the initial CFU. A 10-μl volume of the bacterial stock was incubated in the diluted serum for 30 min with shaking at 200 rpm in a 37°C incubator (Inova 4080; New Brunswick Scientific, Edison, NJ). Serial 1/10 dilutions of the reaction mixtures were diluted into PBSS and were spread on S-BHI with or without appropriate antibiotic selection. These were grown overnight at 37°C under 5% CO2, and emerging colonies were counted the next day. The resulting CFU value was that recorded after 30 min. Killing was assessed by comparing the number of CFU from the 30-min serum incubation with the number of the initial CFU. Results were expressed as the log10 change in CFU at 30 min compared to the initial CFU. Statistical analysis of the data from bactericidal assays was carried out using the paired t test and analysis of variance functions found in GraphPad Prism, version 4 (GraphPad, San Diego, CA).
Biofilm growth in a continuous-flow chamber. NTHi strains 2019 and 2019siaT were grown in a continuous-flow chamber identical to that described by Davies et al. (5). Strains were grown to mid-log phase in RPMI 1640 medium (Gibco BRL, Grand Island, N.Y.) supplemented with protoporphyrin IX (1 μg/ml), hypoxanthine (0.1 mg/ml), uracil (0.1 mg/ml), -NAD (10 μg/ml), sodium pyruvate (0.8 mM), and sialic acid (100 μM). Cultures were diluted to an optical density of 0.25 at 600 nm to inoculate chambers. Chambers were filled with the prepared inoculum and incubated at 37°C for 1 h to allow for adherence of the bacteria to the coverslip. Supplemented RPMI medium was diluted 1:10 with PBS and was pumped into chambers at a flow rate of 180 μl/min. Biofilm was allowed to grow for 2 days. Chambers were photographed and then stained with the LIVE/DEAD BacLight viability stain (Molecular Probes, Eugene, OR) by following the manufacturer's protocol. The biofilm was visualized with a Bio-Rad 1024 laser scanning confocal microscope at a magnification of x20. A 3-dimensional representation of the image was produced using the high-resolution rendering function of the Volocity program (http://www.bucher.ch/index2.html). Image J was used to display the vertical cross section of the biofilms (http://rsb.info.nih.gov/ij/).
Nucleotide sequence accession numbers. GenBank accession numbers are as follows: NTHi 2019 siaT (HI0147), DQ054471; NTHi 3198 HI0147 ORF, DQ054469; NTHi 7502 HI0147 ORF, DQ054470.
RESULTS
Effect of the HI1104 mutation on LOS sialylation. Based on sequence comparison analysis with the E. coli nanT gene sequence, it was assumed that HI1104 encoded a sialic acid transporter. The HI1104 ORF is predicted to encode a protein of 407 amino acids. HI1104 was deleted in the NTHi 2019 strain, and the LOS from the resultant strain was analyzed by SDS-PAGE and, after O deacylation, MALDI-TOF-MS. SDS-PAGE suggested that there was very little difference in sialylation between mutant and wild-type LOS profiles, even when the strains were grown on a medium supplemented with sialic acid (Fig. 1A, compare lanes 1 to 4 with lanes 5 to 8). Treatment of the LOS with neuraminidase showed identical band shifts in the mutant and the wild type, confirming that there are sialylated glycoforms present in the HI1104 mutant (Fig. 1A, compare lanes 2 and 4 with lanes 6 and 8, respectively).
To confirm the SDS-PAGE analysis, we O deacylated the LOS isolated from the wild-type and HI1104 mutant bacteria grown in the presence of sialic acid and subjected the samples to analysis by MALDI-TOF-MS. Previous studies have shown that NTHi 2019 produces a complex mixture of LOS glycoforms (7, 9, 20). The major component of this mixture has been extensively studied and is known to consist of a lactose moiety (Gal1,4-Glc1-) linked in a 1,4 linkage to HepI of the characteristic core structure of H. influenzae [HepIII1,2-HepII1,3-HepI1,5-Kdo(P)-lipid A] (19, 20, 24). The wild-type NTHi 2019 O-LOS (Fig. 1B and Table 2) gave a similar repertoire of glycoforms as that seen previously by MALDI-TOF-MS (9), with the major glycoform being the lactose-containing glycoform modified with 2 or 3 phosphoethanolamine (PEA) moieties (B2 and B3, respectively). Various larger glycoforms are present which differ from the B glycoforms by the addition of as many as four hexoses and a single N-acetylhexosamine. Additionally, many of the glycoforms can be decorated with up to two sialic acid moieties; such glycoforms disappear upon neuraminidase treatment of the O-LOS (data not shown). The MALDI-TOF-MS spectra of the HI1104 mutant (Fig. 1C; Table 2) appear to be very similar to the wild-type spectra, expressing both asialylated and, importantly, sialylated glycoforms. further demonstrating that HI1104 does not function as the sole sialic acid transporter in NTHi 2019. These data suggested that either multiple sialic acid transporters are present in NTHi or this gene does not encode the functional sialic acid transporter of NTHi.
Dose dependence of incorporation of sialic acid into LOS. To study the efficiency of sialic acid uptake in NTHi 2019, the wild type and a mutant lacking the neuraminyl lyase gene (nanA) were grown in defined sialic acid-free medium containing increasing amounts of sialic acid (0 to 1.0 mM). The LOS was then isolated from these bacteria, O deacylated with anhydrous hydrazine, and analyzed by MALDI-TOF-MS. The nanA mutant was included to create a condition where the sole fate of the sialic acid was incorporation into LOS. In the absence of sialic acid, the LOS profiles for the wild type and nanA mutant are almost identical (Fig. 2A). There are differences in the diversity and expression of the glycoforms observed in this experiment compared to the O-LOS in Fig. 1B. These differences are thought to occur due to whether the bacteria grow adhered to a surface (plates) or planktonically (liquid medium) (9). When the bacteria are grown in the presence of 1 mM sialic acid, the spectra of the O-LOS differ significantly (Fig. 2C and D). Although the wild type and the nanA mutant express the same glycoforms, the asialylated B glycoforms no longer predominate as the major LOS glycoform from the nanA mutant; rather, the disialylated B3 glycoform predominates (compare Fig. 2C and D). This confirms the observations of Vimr et al. (28) that the nanA mutant hypersialylates its LOS. When the ion peaks from the spectra are integrated and the sialylated glycoforms are expressed as a percentage of the total glycoforms (Fig. 2E), we see that the wild type is much less sensitive to increases in sialic acid concentration in the medium, suggesting that the nanA sialic acid degradation pathway is in balance with the CMP-sialic acid synthetase "activation" pathway, thus limiting the appearance of sialic acid in the LOS. The nanA mutant likely gives a more accurate indication of the efficiency of sialic acid transport, showing that LOS glycoforms containing sialic acid can be identified at sialic acid concentrations as low as 0.1 μM.
Cloning and mutagenesis of siaT. Figure 3A shows the genomic region surrounding siaT. The siaT ORF from the H. influenzae Rd genome database encodes a predicted protein of 633 amino acids. It was identified by homology with genes known to encode the transmembrane components of TRAP-type C4 dicarboxylate transporters from Vibrio cholerae (VC1777), Vibrio vulnificus (VVA1200), Haemophilus somnus (2336_DctQ), Pasturella multocida (PM1078), and Fusobacterium nucleatum (FN1473) (for sequence alignments, see Fig. 1S in the supplemental material). The siaT gene appears to be a fusion of the two membrane-spanning components of the TRAP-type transport system. The proximity of this gene encoding a potential transporter to the nan operon suggests that this gene may be involved in the uptake of environmental sialic acid. The predicted amino acid sequences of the siaT (HI0147) genes from NTHi 2019, 3198, and 7502 were aligned. The 7502 sequence differs from 2019 only in two amino acids (Ile13Met and Ser29Ala) (see Fig. 1S in the supplemental material). Plasmid p7502HI0147 was used to create the deletion mutant in all NTHi backgrounds, because it was the only plasmid that had convenient restriction enzyme sites near the center of the insert sequence. Transformants of 2019siaT were screened using a colony blot assay with monoclonal antibody 3F11, which is specific for an N-acetyllactosamine epitope (31) that is masked by the addition of sialic acid to the LOS. In the wild type, the 3F11 epitope is observed only upon treatment of the blotted colonies with neuraminidase; however, in the two transformants, 3F11 binding was observed both before and after neuraminidase treatment, suggesting that these mutants lack sialic acid and thus display N-acetyllactosamine as a terminal disaccharide on their LOS. The expression of HI0148, annotated as a hypothetical protein, 168 base pairs upstream of siaT was confirmed by reverse transcription-PCR (data not shown). siaT mutants were also created in NTHi 3198 and 7502 backgrounds.
Uptake of [3H]sialic acid by NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA. NTHi 2019, 2019nanA, 2019siaT, and 2019siaTnanA were studied for their abilities to transport [3H]sialic acid. The 3H label of the sialic acid used in the uptake assay is located on C-9 and thus forms part of the pyruvate when the sialic acid is metabolized by the neuraminyl lyase (NanA), ultimately being lost as 3H2O. Thus, the uptake assay is measuring the incorporation of sialic acid into the LOS and does not take into account the sialic acid that enters the degradation pathway. Again, the neuraminyl lyase (nanA) mutants were included in this study to ensure that all of the transported sialic acid was incorporated into the LOS rather than into degradation pathways (28). Figure 3 shows the results of these studies. As can be seen, 2019siaT and 2019siaTnanA were unable to transport sialic acid and failed to accumulate an appreciable amount of the sialic acid over the 20-min time course of the experiment. During this time, NTHi 2019 acquired 8- to 9-fold more sialic acid than NTHi 2019siaT, while NTHi 2019nanA accumulated 80-fold more sialic acid than 2019siaTnanA. Additionally, the wild type rapidly achieved a steady state (<1 min), whereas the 2019nanA mutant continued to accumulate sialic acid, reaching saturation after 5 min. These studies indicate that the protein encoded by siaT is the sialic acid transporter in NTHi 2019. Uptake studies performed on siaT mutants of NTHi 3198 and NTHi 7502 gave similar results (data not shown).
Comparative analysis of LOS from NTHi 2019 and 2019siaT. LOS from wild-type NTHi 2019 and the siaT mutant were prepared from these strains grown on S-BHI agar with or without supplemental sialic acid. A portion of the LOS sample was treated with neuraminidase, and then pre- and post-neuraminidase treatment samples were resolved by SDS-PAGE (Fig. 4A). The wild-type NTHi 2019 gave a typical glycoform pattern for this strain, with a number of bands that became intensified upon growth of the bacteria with supplemental sialic acid (Fig. 4A, compare lanes 1 and 3). These intensified glycoforms disappeared upon treatment of the LOS with neuraminidase, and the acceptor glycoforms became intensified (Fig. 4A, compare lanes 1 and 3 with 2 and 4, respectively). In contrast, the glycoforms observed for the 2019siaT mutant remained the same regardless of whether or not supplemental sialic acid was added to the growth medium (Fig. 4A, compare lanes 5 and 7). Equally, there were no differences in the LOS profiles after neuraminidase treatment (Fig. 4A, compare lanes 5 and 7 with 6 and 8, respectively). It seems, upon comparison of the LOS of the mutant with that of the wild type, that the profile of the mutant is more similar to that of the neuraminidase-treated wild-type LOS. Taken together, these data suggest that in this mutant, sialic acid is not incorporated into the LOS. Western blot analysis using 3F11 was carried out on the wild-type and 2019siaT mutant LOS. The wild-type LOS grown on BHI (which contains trace amounts of sialic acid) without supplemental sialic acid was negative for 3F11 binding, suggesting that enough sialic acid could be incorporated into the LOS to mask the 3F11 epitope (Fig. 4B, lane 1). This was also the case when the wild-type bacteria were grown on BHI with supplemental sialic acid (Fig. 4B, lane 3). When treated with neuraminidase, wild-type 2019 grown under both conditions showed similar banding patterns, corresponding to LOS glycoforms containing terminal N-acetyllactosamine (Fig. 4B, lanes 2 and 4). For the 2019siaT mutant, bands were detected even when the bacteria were grown with supplemental sialic acid, suggesting an inability to mask this epitope (Fig. 4B, lanes 5 and 7). The number of bands detected after neuraminidase treatment remained the same, showing that no further epitopes were unmasked after this treatment (Fig. 4B, lanes 6 and 8). Interestingly, there were striking differences between the epitopes present in the wild type and the 2019siaT mutant after neuraminidase treatment, suggesting that there may be regulation of which glycoforms are expressed when sialic acid is not available to the bacteria (Fig. 4B, compare lanes 2 and 4 with lanes 6 and 8, respectively).
MALDI-TOF-MS of O-deacylated LOS. To further investigate the LOS phenotype, wild-type and siaT mutant LOS were O deacylated by treatment with anhydrous hydrazine and analyzed by MALDI-TOF-MS. MALDI-TOF-MS of NTHi 2019 (Fig. 5A and Table 2) gave a similar repertoire of glycoforms as that seen in the HI1104 study in Fig. 2A to D. The proportion of sialylated glycoforms present increased when the wild-type bacteria were grown in the presence of sialic acid (Fig. 5B). The MALDI-TOF-MS spectra of the O-LOS from 2019siaT showed a similar diversity in glycoforms; however, the mutant lacked the sialic acid-containing LOS glycoforms that could be seen in the wild type (Fig. 5C). These sialylated species were also completely absent from the O-LOS of 2019siaT grown on sialic acid-supplemented medium (Fig. 5D), indicating that in the absence of siaT, the bacteria were not capable of incorporating sialic acid into their LOS.
Bactericidal assay. It has been shown previously that there is a correlation between the incorporation of terminal sialic acid into LOS and protection of H. influenzae from complement-mediated killing by normal human serum. Since 2019siaT cannot acquire sialic acid for incorporation into LOS, it is likely that such a bacterium would have an increased susceptibility to killing by normal human serum. To investigate this, we carried out a bactericidal assay on the wild type, 2019siaT, 2019nanA, and 2019siaTnanA. In the absence of supplemental sialic acid in the medium, the wild type, 2019siaT, and 2019siaTnanA were all susceptible to killing by normal human serum (Fig. 6A). Interestingly, the 2019nanA mutant was resistant to serum killing, suggesting that the bacterium can acquire sufficient sialic acid from the BHI medium (which contains trace amounts of sialic acid) to afford protection from serum killing. The wild type and 2019nanA grown in a medium supplemented with 20 μM sialic acid were protected from serum killing by incorporation of sialic acid into their LOS (Fig. 6B). Conversely, 2019siaT and 2019siaTnanA, when grown in the presence of sialic acid, were still susceptible to serum killing (Fig. 6B). As a control, the experiment was repeated using heat-inactivated serum; as expected, both the wild type and mutants were capable of surviving this treatment (Fig. 6C and D). This finding clearly supports the MALDI-TOF-MS data and shows that in the absence of siaT, H. influenzae is incapable of incorporating sialic acid into its LOS.
Biofilm formation. Previous work has indicated that sialic acid is a significant component of the biofilm formed by NTHi 2019 (9). A continuous-flow cell assay was used to determine the ability of the 2019siaT mutant to form a biofilm. After 2 days of growth, both strains produced a visible biofilm (Fig. 7A and B). LIVE/DEAD staining revealed that a majority of cells in the wild-type biofilm were viable, while the 2019siaT biofilm had a higher proportion of dead cells. Additionally, the 2019siaT biofilm appeared to have a lower cell density than the wild-type biofilm, as judged by the signal intensity of the Syto-9 dye. While inactivation of siaT does not inhibit biofilm formation, the loss of sialic acid transport clearly has a profound effect on the viability of cells in the biofilm and the superstructure of the biofilm.
DISCUSSION
The surface of nontypeable Haemophilus influenzae is covered with LOS molecules. Incorporation of terminal sialic acid into the LOS enables NTHi to evade complement-dependent killing mechanisms, an important component of the host immune system (10, 11). The pathways that are involved in the incorporation of sialic acid into the LOS and/or its regulation are important targets for future drug development. Many of the genes involved in the metabolism of sialic acid in H. influenzae have been characterized, including those encoding the CMP-sialic acid synthetase (11), sialyltransferases (10, 12), and the neuraminyl lyase (28). As yet the sialic acid transporter of H. influenzae has not been characterized. Until recently it was thought that the sialic acid transporter of H. influenzae was encoded by the HI1104 ORF, a gene that has homology to the nanT gene of E. coli (30). The nanT gene product is a secondary transporter of the major facilitator superfamily that imports sialic acid in symport with a proton. Deletion of the HI1104 nanT homolog in NTHi 2019 had little effect on the ability of the bacteria to incorporate sialic acid into their LOS. H. influenzae lacks a sialic acid synthesis pathway, suggesting that either HI1104 is not a sialic acid transporter in NTHi 2019 or there is a second transporter that is capable of importing sialic acid into the bacteria.
To gain further insight into the efficiency of sialic acid uptake by NTHi 2019, the dose dependence of incorporation of sialic acid into the LOS was investigated. The process appears to be highly efficient in the nanA mutant background, creating a condition where all sialic acid is directed toward LOS sialylation; sialic acid can be detected in the LOS when the concentration of sialic acid in the medium is as low as 0.1 μM. More recently, it has been suggested in the literature that H. influenzae has a sialic acid transporter which is different from that of E. coli nanT (30). The genes HI0146 and HI0147 (siaT) have a close proximity in the H. influenzae genome to the nan operon, which encodes the genes responsible for the catabolism of sialic acid. By homology, these genes were predicted to encode an ESR and the transporter domains, respectively, of a transporter of the TRAP family (14, 21). Such transporters have been reported previously in bacteria and are often involved in the transport of C4 dicarboxylates such as succinate and fumarate using the electrochemical proton gradient as a driving force (6). To investigate the hypothesis that siaT acts as a sialic acid transporter in NTHi, the siaT gene was deleted in NTHi strains 2019, 3198, and 7502.
The sialic acid uptake assay demonstrates that the siaT gene product is likely the sialic acid transporter of NTHi. The 2019siaT and 2019siaTnanA mutants were both incapable of sialic acid uptake, indicating that the siaT gene product is required prior to activation of sialic acid by the CMP-sialic acid synthetase, thus implicating it as the transporter in NTHi 2019. The same assay repeated on siaT mutants in NTHi strains 3198 and 7502 gave similar results (data not shown). The assay also demonstrates some interesting aspects of the uptake and incorporation of sialic acid into the LOS of H. influenzae. In the wild type, the amount of sialic acid detected rapidly reaches a "steady-state" level, in contrast to the 2019nanA mutant, which continues to accumulate sialic acid until reaching saturation. Indeed, it has been shown previously that H. influenzae nanA mutants hypersialylate their LOS (28). This would suggest that the point of regulation of sialic acid levels within the bacteria occurs not through regulation of the transporter but through regulation of the downstream gene products in the sialic acid pathway such as the neuraminyl lyase and the CMP-N-sialic acid synthetase. The concept that the sialic acid transporter is expressed at a constitutive level would make some sense, because sialic acid, as well as being an important molecule in the evasion of the host immune response, is a valuable carbon and nitrogen source.
Deletion of siaT, as suggested by the sialic acid uptake data, results in bacteria that are incapable of incorporating sialic acid into their LOS, thus suggesting that this gene is indeed involved in the uptake of sialic acid in NTHi. This conclusion is supported by the SDS-PAGE and MALDI-TOF-MS data, all of which conclusively indicate the lack of sialic acid-containing LOS glycoforms on the 2019siaT mutant. As a consequence, the siaT mutant is severely compromised in its ability to evade the host immune response, as evidenced by the fact that the mutant is susceptible to complement-mediated killing when exposed to normal human sera. The Western blot analysis suggested that the 2019siaT mutant was not only compromised in its ability to make sialylated LOS glycoforms but also expressed fewer glycoforms containing terminal N-acetyllactosamine. This suggests that in the absence of sialic acid, the bacteria express different LOS glycoforms, although the mechanism of this regulation remains to be investigated.
A point of secondary interest is that the 2019nanA mutant, when grown on BHI medium (which contains trace amounts of sialic acid), can acquire enough sialic acid from the medium to produce LOS that is sufficiently sialylated to protect the bacterium from complement-mediated lysis, thus indicating that the deletion of nanA leads to an increase in "flux" of sialic acid into the LOS incorporation pathway.
Finally, in the absence of Neu5Ac transport, NTHi 2019siaT, although capable of forming a biofilm, produces a biofilm with altered morphology, lower cell density, and a higher proportion of dead cells than wild-type NTHi 2019 biofilm. We are uncertain why bacterial viability is reduced when Neu5Ac is limiting. A similar phenotype was seen with 2019siaA during middle ear infection in a chinchilla model of otitis media (13).
This is the first TRAP transporter to be characterized in H. influenzae and also the first TRAP transporter known to transport sialic acid. The best-described and -characterized TRAP transporter is the DctPQM C4 dicarboxylate transporter of Rhodobacter capsulatus, though various homologs have been identified in archaea and gram-negative bacteria. DctP is the periplasmic extracellular solute receptor, while DctQ and DctM represent the integral membrane proteins with 4 and 12 predicted membrane-spanning regions, respectively (6). In H. influenzae these two membrane proteins are encoded by a single gene containing a total of 16 predicted membrane-spanning regions (21). The ESR protein is thought to increase the uptake affinity of the transporter by binding sialic acid and delivering it to the transporter. Such a high-affinity sialic acid uptake system may be important for the bacteria in their normal physiological environment. The novelty of this transporter may make it an important drug target given the dependence of H. influenzae on sialic acid for immune evasion.
The diversity that exists among sialic acid transporters in gram-negative bacteria, which so far include members of the major facilitator family and TRAP transporters, is noteworthy. This diversity may be driven, in part, by the particular environments to which bacteria have adapted, the challenges these environments present for their growth and survival, and the abundance of sialic acid that is present in these environments. For example, although the nasopharynx is rich in secreted sialylated mucins and possibly other types of sialoglycoconjugates, free sialic acid is much more limited, and the mechanism by which H. influenzae may exploit these more abundant bound sources of sialic acid remains to be elucidated.
In conclusion, deletion of the siaT gene of H. influenzae leads to bacteria that are not capable of sialic acid uptake; thus, the bacteria cannot sialylate their LOS, making them vulnerable to complement-mediated killing. The siaT gene product therefore appears to be the sole sialic acid transporter of the NTHi strains investigated. The siaT gene product is a transporter of the TRAP transporter family, making this the first sialic acid transporter of this type to be investigated.
ACKNOWLEDGMENTS
This work was supported by grants from the National Institutes of Health (AI24616 to M.A.A. and B.W.G.; AI30040 to M.A.A.).
Supplemental material for this article may be found at http://iai.asm.org/.
REFERENCES
1. Andreoni, J., H. Kayhty, and P. Densen. 1993. Vaccination and the role of capsular polysaccharide antibody in prevention of recurrent meningococcal disease in late complement component-deficient individuals. J. Infect. Dis. 168:227-231.
2. Campagnari, A. A., M. R. Gupta, K. C. Dudas, T. F. Murphy, and M. A. Apicella. 1987. Antigenic diversity of lipooligosaccharides of nontypeable Haemophilus influenzae. Infect. Immun. 55:882-887.
3. Coleman, H. N., D. A. Daines, J. Jarisch, and A. L. Smith. 2003. Chemically defined media for growth of Haemophilus influenzae strains. J. Clin. Microbiol. 41:4408-4410.
4. Davidson, A. L., and J. Chen. 2004. ATP-binding cassette transporters in bacteria. Annu. Rev. Biochem. 73:241-268.
5. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298.
6. Forward, J. A., M. C. Behrendt, N. R. Wyborn, R. Cross, and D. J. Kelly. 1997. TRAP transporters: a new family of periplasmic solute transport systems encoded by the dctPQM genes of Rhodobacter capsulatus and by homologs in diverse gram-negative bacteria. J. Bacteriol. 179:5482-5493.
7. Gaucher, S. P., M. T. Cancilla, N. J. Phillips, B. W. Gibson, and J. A. Leary. 2000. Mass spectral characterization of lipooligosaccharides from Haemophilus influenzae 2019. Biochemistry 39:12406-12414.
8. Gibson, B. W., J. J. Engstrom, C. M. John, W. Hines, and A. M. Falick. 1997. Characterization of bacterial lipooligosaccharides by delayed extraction matrix-assisted laser desorption time-of-flight mass spectrometry. J. Am. Soc. Mass Spectrom. 8:645-658.
9. Greiner, L. L., H. Watanabe, N. J. Phillips, J. Shao, A. Morgan, A. Zaleski, B. W. Gibson, and M. A. Apicella. 2004. Nontypeable Haemophilus influenzae strain 2019 produces a biofilm containing N-acetylneuraminic acid that may mimic sialylated O-linked glycans. Infect. Immun. 72:4249-4260.
10. Hood, D. W., A. D. Cox, M. Gilbert, K. Makepeace, S. Walsh, M. E. Deadman, A. Cody, A. Martin, M. Mansson, E. K. Schweda, J. R. Brisson, J. C. Richards, E. R. Moxon, and W. W. Wakarchuk. 2001. Identification of a lipopolysaccharide -2,3-sialyltransferase from Haemophilus influenzae. Mol. Microbiol. 39:341-350.
11. Hood, D. W., K. Makepeace, M. E. Deadman, R. F. Rest, P. Thibault, A. Martin, J. C. Richards, and E. R. Moxon. 1999. Sialic acid in the lipopolysaccharide of Haemophilus influenzae: strain distribution, influence on serum resistance and structural characterization. Mol. Microbiol. 33:679-692.
12. Jones, P. A., N. M. Samuels, N. J. Phillips, R. S. Munson, Jr., J. A. Bozue, J. A. Arseneau, W. A. Nichols, A. Zaleski, B. W. Gibson, and M. A. Apicella. 2002. Haemophilus influenzae type b strain A2 has multiple sialyltransferases involved in lipooligosaccharide sialylation. J. Biol. Chem. 277:14598-14611.
13. Jurcisek, J., L. Greiner, H. Watanabe, A. Zaleski, M. A. Apicella, and L. O. Bakaletz. 2005. Role of sialic acid and complex carbohydrate biosynthesis in biofilm formation by nontypeable Haemophilus influenzae in the chinchilla middle ear. Infect. Immun. 73:3210-3218.
14. Kelly, D. J., and G. H. Thomas. 2001. The tripartite ATP-independent periplasmic (TRAP) transporters of bacteria and archaea. FEMS Microbiol. Rev. 25:405-424.
15. Kolker, E., K. S. Makarova, S. Shabalina, A. F. Picone, S. Purvine, T. Holzman, T. Cherny, D. Armbruster, R. S. Munson, Jr., G. Kolesov, D. Frishman, and M. Y. Galperin. 2004. Identification and functional analysis of ‘hypothetical’ genes expressed in Haemophilus influenzae. Nucleic Acids Res. 32:2353-2361.
16. Lesse, A. J., A. A. Campagnari, W. E. Bittner, and M. A. Apicella. 1990. Increased resolution of lipopolysaccharides and lipooligosaccharides utilizing tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis. J. Immunol. Methods 126:109-117.
17. Mandrell, R. E., R. McLaughlin, Y. Aba Kwaik, A. Lesse, R. Yamasaki, B. Gibson, S. M. Spinola, and M. A. Apicella. 1992. Lipooligosaccharides (LOS) of some Haemophilus species mimic human glycosphingolipids, and some LOS are sialylated. Infect. Immun. 60:1322-1328.
18. Mansson, M., S. H. Bauer, D. W. Hood, J. C. Richards, E. R. Moxon, and E. K. Schweda. 2001. A new structural type for Haemophilus influenzae lipopolysaccharide. Structural analysis of the lipopolysaccharide from nontypeable Haemophilus influenzae strain 486. Eur. J. Biochem. 268:2148-2159.
19. Masoud, H., E. R. Moxon, A. Martin, D. Krajcarski, and J. C. Richards. 1997. Structure of the variable and conserved lipopolysaccharide oligosaccharide epitopes expressed by Haemophilus influenzae serotype b strain Eagan. Biochemistry 36:2091-2103.
20. Phillips, N. J., M. A. Apicella, J. M. Griffiss, and B. W. Gibson. 1992. Structural characterization of the cell surface lipooligosaccharides from a nontypable strain of Haemophilus influenzae. Biochemistry 31:4515-4526.
21. Rabus, R., D. L. Jack, D. J. Kelly, and M. H. Saier, Jr. 1999. TRAP transporters: an ancient family of extracytoplasmic solute-receptor-dependent secondary active transporters. Microbiology 145:3431-3445.
22. Rao, V. K., G. P. Krasan, D. R. Hendrixson, S. Dawid, and J. W. St Geme III. 1999. Molecular determinants of the pathogenesis of disease due to non-typable Haemophilus influenzae. FEMS Microbiol. Rev. 23:99-129.
23. Schneider, H., J. M. Griffiss, J. W. Boslego, P. J. Hitchcock, K. M. Zahos, and M. A. Apicella. 1991. Expression of paragloboside-like lipooligosaccharides may be a necessary component of gonococcal pathogenesis in men. J. Exp. Med. 174:1601-1605.
24. Schweda, E. K., O. E. Hegedus, S. Borrelli, A. A. Lindberg, J. N. Weiser, D. J. Maskell, and E. R. Moxon. 1993. Structural studies of the saccharide part of the cell envelope lipopolysaccharide from Haemophilus influenzae strain AH1-3 (lic3+). Carbohydr. Res. 246:319-330.
25. Teele, D. W., J. O. Klein, B. Rosner, L. Bratton, G. R. Fisch, O. R. Mathieu, P. J. Porter, S. G. Starobin, L. D. Tarlin, and R. P. Younes. 1983. Middle ear disease and the practice of pediatrics. Burden during the first five years of life. JAMA 249:1026-1029.
26. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.
27. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 119:115-119.
28. Vimr, E., C. Lichtensteiger, and S. Steenbergen. 2000. Sialic acid metabolism's dual function in Haemophilus influenzae. Mol. Microbiol. 36:1113-1123.
29. Vimr, E. R., K. A. Kalivoda, E. L. Deszo, and S. M. Steenbergen. 2004. Diversity of microbial sialic acid metabolism. Microbiol. Mol. Biol. Rev. 68:132-153.
30. Vimr, E. R., and F. A. Troy. 1985. Identification of an inducible catabolic system for sialic acids (nan) in Escherichia coli. J. Bacteriol. 164:845-853.
31. Yamasaki, R., W. Nasholds, H. Schneider, and M. A. Apicella. 1991. Epitope expression and partial structural characterization of F62 lipooligosaccharide (LOS) of Neisseria gonorrhoeae: IgM monoclonal antibodies (3F11 and 1-1-M) recognize non-reducing termini of the LOS components. Mol. Immunol. 28:1233-1242.(Simon Allen, Anthony Zale)