Abstract
We have determined the sequence of the gene cluster encoding the O antigen in Francisella novicida and compared it to the previously reported O-antigen cluster in Francisella tularensis subsp. tularensis. Immunization with purified lipopolysaccharide (LPS) from F. tularensis subsp. tularensis or F. novicida protected against challenge with Francisella tularensis subsp. holarctica and F. novicida, respectively. The LPS from F. tularensis subsp. tularensis did not confer protection against challenge with F. novicida, and the LPS from F. novicida did not confer protection against challenge with F. tularensis subsp. holarctica. Allelic replacement mutants of F. tularensis subsp. tularensis or F. novicida which failed to produce O antigen were attenuated, but exposure to these mutants did not induce a protective immune response. The O antigen of F. tularensis subsp. tularensis appeared to be important for intracellular survival whereas the O antigen of F. novicida appeared to be critical for serum resistance and less important for intracellular survival.
Francisella tularensis is the causative agent of tularemia, a disease that affects many mammals including humans and rodents. The bacterium is a small (0.2 to 0.5 μm by 0.7 to 1.0 μm) gram-negative intracellular pathogen, and the natural reservoir is thought to be rodents, with ticks being the primary vector (11).
F. tularensis has been divided into three subspecies: F. tularensis subsp. tularensis (also known as type A) is the most virulent and was thought to be confined to North America but has recently been isolated in Europe (16, 26). F. tularensis subsp. holarctica (also known as type B) is less virulent and is found mainly in Europe and in Asia (27). F. tularensis subsp. mediasiatica has only been isolated from locations in Central Asia (27, 32) and is considered to be of relatively low virulence (10, 26). In addition, Francisella novicida, which is considered by some investigators to be another subspecies of F. tularensis, is reported to be pathogenic only in immunocompromised humans (10).
The sequences of O-antigen gene clusters and the structures of the O antigens from strain SchuS4 (F. tularensis subsp. tularensis) and from the LVS strain (F. tularensis subsp. holarctica) have previously been shown to be identical (31, 39, 40). In contrast, although the F. novicida (strain U112) O antigen has many sugars in common with the F. tularensis subsp. tularensis and F. tularensis subsp. holarctica O antigens, it is structurally distinct (39). This suggests that the O-antigen gene cluster is also distinct in F. novicida. To date, the sequences of the genes that encode the O antigen of the LPS of F. novicida have not been reported.
Several studies have shown that antibody to the lipopolysaccharide (LPS) of F. tularensis can play a role in protective immunity to tularemia (12, 31). It is not clear whether the F. novicida O antigen is immunologically distinct from the O antigen produced by F. tularensis subsp. tularensis and F. tularensis subsp. holarctica. However, Shen et al. showed that immunization of mice with F. novicida induces only marginal protection against a subsequent challenge with F. tularensis subsp. tularensis or F. tularensis subsp. holarctica strains (35). This suggests that the differences in the structures of the O antigens from F. novicida and F. tularensis subsp. tularensis and F. tularensis subsp. holarctica structure are too great to confer cross-protection.
There is some evidence from studies with the LVS strain that the O antigen plays a role in the virulence of F. tularensis and F. novicida (17). The LVS strain forms two colony types or opacity variants, referred to as blue and gray variants, when grown on solid medium (9). The gray variants are devoid of O antigen and are less virulent than the blue variants. In addition, they do not induce protective immunity in mice (17). However, the genetic events which accompany this phenotypic change are not known, and it is possible that there are differences additional to the loss of O-antigen expression (17). Also, it is known that the pathogenesis of tularemia in mice caused by the LVS strain is not identical to the pathogenesis of the disease caused by high-virulence strains (5, 7). Therefore, these studies may not indicate the roles of the O antigen in highly virulent strains of F. tularensis.
The aim of this study was to compare and contrast the roles of the O antigens in virulence of F. tularensis subsp. tularensis and F. novicida and to determine the extent of immunological cross-reactivity between these antigens.
MATERIALS AND METHODS
Bacteria, plasmids, growth conditions, and general enzymes and chemicals.
The bacterial strains and plasmids used in this study are listed in Table 1. Unless otherwise stated, enzymes for the manipulation of DNA, nucleotides, and reagents for the detection of the digoxigenin (DIG)-labeled probe for Southern blotting were obtained from Roche Diagnostics Limited (Lewes, United Kingdom); chemicals were obtained from Sigma Chemical Co. (Poole, United Kingdom), and culture medium was obtained from Oxoid Limited (Basingstoke, United Kingdom).
TABLE 1.
Bacterial strains and plasmids used in this study
| Strain or plasmid | Characteristics or descriptiona | Reference or sourceb |
|---|---|---|
| Strains | ||
| E. coli S17-1 | thi thr leu tonA lacY supE recA::RP4-2-Tc::Mu Kn::Tn7 | 36 |
| F. novicida strain U112 | Avirulent strain | Water, 1950, Utah (16, 19) |
| F. tularensis subsp. tularensis SchuS4 | Virulent type A | Human ulcer, 1941, Ohio; FSC237 |
| F. tularensis subsp. holarctica HN63 | Virulent type B | Hare, 1963, Telemark, Norway; FSC092 |
| Plasmids | ||
| pSMP22 | Suicide vector for allelic replacements in F. tularensis, constructed based on pPV2 (a gift from I. Golovliov) (15). CHL resistance gene was removed by digestion with HindIII, and the plasmid backbone was religated. A F. tularensis groES promoter with sacB fusion cassette was cloned into the SalI/MluI sites. | This study |
| pSMP22-RT1 | Suicide plasmid used to inactivate O-antigen cluster of F. tularensis subsp. tularensis; 2.5-kb fragment cloned into pSMP22; CHL resistance cassette inserted between amplified fragments of F. tularensis subsp. tularensis wbtD and wbtF genes. | This study |
| pSMP22-RT2 | Suicide plasmid used to inactivate O-antigen cluster of F. novicida; 2.5-kb fragment cloned into pSMP22; CHL resistance cassette inserted between amplified fragments of F. novicida wbtDFn and wbtFFn genes. | This study |
CHL, chloramphenicol.
Identifiers are from the Francisella Strain Collection (FSC) in Umeå, Sweden.
Strains of F. tularensis subsp. tularensis and F. novicida (19, 22) were cultured on blood cysteine glucose agar (BCGA) supplemented with 10 ml of 10% (wt/vol) histidine per liter or on modified Thayer Martin agar (BBL GC agar base[Becton Dickinson, United Kingdom] supplemented with 1% [wt/vol] hemoglobin and 1% [vol/vol] BBL Isovitalex [Becton Dickinson]) or in liquid culture in Chamberlain's Defined Medium (CDM) (4). Escherichia coli was cultured on Luria-Bertani (LB) plates or broth (3). Chloramphenicol was added to growth medium at 25 μg/ml for E. coli and 10 μg/ml for F. novicida and F. tularensis subsp. tularensis.
Identification and sequencing of the O-antigen gene cluster of F. novicida.
Presence of SchuS4-like O-antigen genes in F. novicida strain U112 was determined by PCR. Primers were designed from the SchuS4 sequence. Primers IsFtu2 F, 2F, 2R, 3F, wbtF F, wbtG R, wbtG F, wbtH F, wbtH R, and wbtMR (Table 2) were used in combination to sequence the O-antigen gene cluster using the Expand Long Template polymerase (Roche), according to the manufacturer's instructions. PCR products were sequenced by primer walking to publication quality by MWG (Ebersberg, Germany). Genes amplified from F. novicida encoding proteins with high homologies to O-antigen proteins of F. tularensis subsp. tularensis were denoted with the subscript Fn. Those genes which were found to have no homology to O-antigen genes of F. tularensis subsp. tularensis were instead given a new gene name.
TABLE 2.
Oligonucleotides used in this study
| Name | Sequence (5′-3′)a | Purpose | Location relative to SchuS4 genome (bp) |
|---|---|---|---|
| wbtD F | ACGCGTCGACTCCAAATGTAACAGGCTTAG | Disruption cassette cloning; used to make Southern blot probe | 1512972-1512953 |
| wbtD R | GTGAAAGTTGGAACCTCTTACAGATCTACGTGCATTTTGCTGTAAGTATGCATTGAC | Disruption cassette cloning; used to make Southern blot probe | 1512219-1512242 |
| wbtF F | GATGAGTGGCAGGGCGGGGCGTAAAGATCTTGCACGGCTTACGATAATGTTAAATTTC | Disruption cassette cloning | 1510886-1510864 |
| wbtF R | ACGCGTCGACATTAATTAAATACCACTCAACAGC | Disruption cassette cloning | 1509920-1509943 |
| wbtE F | TAGTAGTAGACGCAGGAG | Used to make Southern blot probe | 1511621-1511604 |
| wbtE R | AGCTGCCTTGTACGTTAG | Used to make Southern blot probe | 1511365-1511382 |
| Cam F | GCACGTAGATCTGTAAGAGGTTCCAACTTTCAC | Used to make Southern blot probe | |
| Cam R | CGTGCAAGATCTTTACGCCCCGCCCTGCCACTCATC | Used to make Southern blot probe | |
| wbtF F | CTAATGAAGAGTGGGCGAGAC | PCR | 1510407-1510387 |
| wbtF R | TATCACCCGCTCTATCTGGAC | PCR | 1510030-1510050 |
| wbtG F | CCTTAACCAAGGTGGTGCTGAA | PCR | 1508645-1508624 |
| wbtG R | ACTCCCCTACCCATAAGTGA | PCR | 1508544-1508563 |
| wbtH F | CTGAGCATGCAAGAGCAGTAG | PCR | 1506655-1546635 |
| wbtH R | ATCACCAGCGTCACCTGATAG | PCR | 1506450-1506470 |
| wbtM R | GCATAACGCCAATCATGTCC | PCR | 1500120-1500074 |
| 2F | TGTCTTAGATATGGGGCAACC | PCR | 1515021-1515001 |
| 2R | ACAAATATCAAATCCTAACACATC | PCR | 1512073-1512096 |
| 3F | TAGAAGCAGCTGCGATAGGTAGAC | PCR | 1512463-1512440 |
| ISFtu2 F | ATGAATTATCATATAAAAGAAGTATTC | PCR | 1517296-1517250 |
| GroES prom F | GTACGTGCACAATAAACATCGCAAAAGGTGTA | Construction of pSMP22 | 1763632-1763653 |
| GroES prom R | ACTTGATCATCGGTACCCATAACAATCTTACTCCT | Construction of pSMP22 | 1763983-1763966 |
| SacB F | TACCGTACGATGAACATCAAAAAGTTTGCA | Construction of pSMP22; amplification of sacB from pRL271 (GenBank accession no. L05081) | |
| SacB R | TGCCGTACGTTATTTGTTAACTGTTAATTG | Construction of pSMP22; amplification of sacB from pRL271 (GenBank accession L05081) |
SalI sites are in boldface; cat sequences are underlined.
Construction of the O-antigen disruption cassettes.
A region of the F. novicida and F. tularensis subsp. tularensis O-antigen gene clusters (wbtDEFFn and wbtDEF, respectively) was targeted for deletion and the insertion of an antibiotic marker to disrupt the entire cluster's function. An internal region of genes wbtDFn and wbtD (the upstream region) and an internal region of genes wbtFFn and wbtF (the downstream region) were amplified by the PCR. The disruption cassette was then constructed by performing splice overhang extension PCR (6) using a PCR-amplified chloramphenicol acetyltransferase gene (cat). The 2.5-kb DNA fragments were ligated into pSMP22 and transformed into E. coli S17-1 (36). The appropriate plasmids were transferred to F. tularensis subsp. tularensis (pSMP22-RT1) and F. novicida (pSMP22-RT2) by conjugation.
The use of chloramphenicol as a marker has been approved for this specific study. All work undertaken in the United Kingdom is reviewed locally by a genetic manipulation safety committee on a project-by-project basis and is also approved nationally by our health and safety authorities.
Conjugal transfer of pSMP22-RT1 and pSMP22-RT2.
Clones pSMP22-RT1 and pSMP22-RT2 were introduced into F. tularensis subsp. tularensis strain SchuS4 F. novicida by conjugation using essentially the same method as Golovliov et al. (15). Conjugation mixtures were incubated on BCGA plates at 25°C before selection of transconjugants. The transconjugants were selected on modified Thayer Martin plates containing polymyxin B (100 μg/ml for counter-selection of the E. coli strain) and 10 μg/ml chloramphenicol. The plates were incubated at 37°C for up to 5 days.
Analysis of transconjugants.
Southern blotting and PCR were used to confirm the deletion/insertion in the wbtDEF or wbtDEFFn gene region. The F. tularensis subsp. tularensis and the F. novicida double recombinant strains were designated ΔwbtDEFFt::Cm or ΔwbtDEFFn::Cm, respectively.
LPS purification, gel electrophoresis and immunodetection.
The hot phenol-water extraction method was used to purify LPS from 1.16 g of freeze-dried F. novicida (31). An identical extraction method was used to isolate LPS from 2.2 g of F. tularensis subspecies tularensis. Glycine gel electrophoresis was performed using a 12.5% separating gel with a 4.5% stacking gel (21). Colonies were boiled in solubilization buffer prior to incubation with 3.3 mg/ml proteinase K for 1 h at 60°C before being loaded onto gels. For immunodetection a murine monoclonal antibody to F. tularensis LVS LPS O side chain (18) and an antibody to F. novicida LPS (a kind gift from F. Nano) (8) were used. An antispecies horseradish peroxidase-labeled antibody (Amersham, United Kingdom) was used as a secondary antibody. The ECL Plus substrate (GE Healthcare, United Kingdom) was used to detect antibody binding according to the manufacturer's instructions.
Structural analysis of the polysaccharide O antigen: NMR spectroscopy.
The water soluble O-antigen polysaccharide component was isolated from the LPS by acid hydrolysis as outlined by Phillips et al. (29). The lyophilized crude polysaccharide was purified using size exclusion chromatography on a Sephadex G-50 column (15 by 5 by 500 mm), with 0.05 M pyridinium acetate buffer as eluant (pH 4.5), at a flow rate of 16 ml/h. Fractions were assessed for carbohydrate content by charring on thin-layer chromatography plates (H2SO4 and ethanol); those containing carbohydrate were analyzed by nuclear magnetic resonance (NMR) spectroscopy.
Purified polysaccharide was dissolved in deuterium oxide (D2O) (99.998% purity; Cambridge Isotope Laboratories) and cycled through three steps of lyophilization/dissolution to remove exchangeable protons. 1H and 13C NMR experiments were performed at 600 MHz and 150 MHz, respectively, at 298 K in D2O using a Bruker Avance spectrometer. Spectral assignment was verified by the recording of one-dimensional (1D) 1H, correlated spectroscopy, total correlated spectroscopy (120 ms mixing time), and 1H-13C-heteronuclear single-quantum coherence spectra. A sample was also prepared in dimethyl sulfoxide-d6 (99.9% purity; Cambridge Isotope Laboratories) and nuclear Overhauser effect spectroscopy (200 ms) was acquired at 323 K. All spectra were acquired using unmodified pulse sequences from the Bruker pulse sequence library.
Isotyping enzyme-linked immunosorbent assay.
Plates were coated with heat-killed whole bacteria. Anti-mouse Fab immunoglobulin G (IgG) and an anti-mouse μ-chain with specific affinity for the IgM subtypes were used. The antibodies were serially diluted across the plate, and naïve mouse serum was included as a control. The appropriate secondary conjugate was diluted and used, and the reaction was developed with TMB (3,3′,5,5′tetramethyl benzidine) substrate. The plates were then incubated at room temperature, and the reaction was stopped by adding 2 M H2SO4. The absorbance was read at 450 nm. The revelations program (version 4.22) was used to calculate the standard curve and antibody concentration values.
Immunization with LPS and protection studies.
Groups of 6 adult female BALB/c mice (Charles River Laboratories, Kent, United Kingdom) aged 6 to 8 weeks were immunized via the intraperitoneal route with 50 μg of purified LPS in 100 μl phosphate-buffered saline, from either F. tularensis subsp. tularensis or F. novicida. Mice were given three doses of LPS at 14-day intervals. Serum samples were taken from mice 14 days after the final immunization, and the level of antibody against immobilized bacteria was determined. Twenty-one days after immunization the mice were challenged with either 100 times the minimum lethal dose (100 MLD) of F. novicida strain U112 by the intraperitoneal route or 100 MLD of F. tularensis subsp. holarctica strain HN63 by the subcutaneous route. All procedures were carried out in accordance with Home Office guidelines. Animals were observed for signs to indicate humane end points, and these were used to direct culling.
Immunization with ΔwbtDEFFt::Cm or ΔwbtDEFFn::Cm and challenge of mice.
Groups of 5 adult female BALB/c mice aged 6 to 8 weeks were immunized subcutaneously with 105 CFU of ΔwbtDEFFt::Cm. Groups of 6 female BALB/c mice aged 6 to 8 weeks were immunized with 105 CFU of ΔwbtDEFFn::Cm via the subcutaneous route. The bacteria were harvested after 18 h of growth at 37°C on BCGA agar. Mice were challenged on day 49 postvaccination with either 100 MLD of F. novicida strain U112 via the intraperitoneal route or 100 MLD of F. tularensis subsp. tularensis strain SchuS4 via the subcutaneous route. All procedures were carried out in accordance with Home Office guidelines. Animals were observed for signs to indicate humane end points, and these were used to direct culling.
Macrophage infection and survival assays.
The murine macrophage cell line J774A.1 (ECCAC, Salisbury, United Kingdom) was cultivated at 37°C with 5% CO2 in Dulbecco's Modified Essential Medium (Gibco Invitrogen, Paisley, United Kingdom) supplemented with 10% heat-inactivated fetal calf serum (Gibco Invitrogen) and 4 mM l-glutamine. The assay was carried out essentially as described by Hartley et al. with a multiplicity of infection of bacteria to macrophage of 10:1 (17). Values for all time points were determined in duplicate, and each experiment was done in triplicate. Results were plotted as number of CFU/ml.
Serum sensitivity assays.
Bacteria were grown overnight in CDM. Cultures were diluted 1/100 in fresh CDM to determine the MIC using guinea pig serum at final concentrations of 50 to 0.02% (vol/vol) in 96-well polypropylene microtiter plates. The MIC50 was determined as the concentration of the compound that gave a 50% optical density reading compared to wells with no substrate after incubation overnight for F. novicida and after incubation for 48 h for F. tularensis subsp. tularensis at 37°C. These assays were performed in triplicate.
Nucleotide sequence accession number.
The nucleotide sequence of the F. novicida O-antigen gene cluster was determined and deposited in the EMBL/GenBank/DDBJ databases under accession number EF059983.
RESULTS
The F. novicida strain U112 O-antigen biosynthetic gene cluster shows homology with the gene cluster from F. tularensis subsp. tularensis strain SchuS4.
We first sequenced the O-antigen gene cluster in F. novicida strain U112. The cluster was approximately 16 kb in length, contained 12 genes (Table 3), and was flanked by transposases (Fig. 1). The overall G+C content of this region (30.45%) was lower than that reported for F. tularensis subsp. tularensis strain SchuS4 (31) and lower than the reported overall G+C content of the F. novicida genome of 32% (http://www.genome.washington.edu/UWGC/Projects/index.cfm?PID = 168&ST = 4). Based on homologies with gene sequences deposited in the GenBank database, a function for each gene product was assigned. Genes from F. novicida encoding proteins with high homologies to corresponding proteins of F. tularensis subsp. tularensis were denoted with the subscript Fn. The O-antigen repeat unit of F. novicida is shown in Fig. 2, with the putative role of each gene of the O-antigen biosynthetic gene cluster annotated. The O-antigen unit F. novicida contains four sugar residues: three 2-acetamido-2-deoxygalacturonamide (GalNAcAN) residues and one 4,6-diacetamido-2,4,6-dideoxyglucose (Qui2NAc4NAc) residue (39). The biosynthesis of GalNAcAN is likely to involve WbtEFn and WbtFFn as they share 99% homology to proteins WbtF and WbtE (present in the F. tularensis subsp. tularensis O-antigen cluster) and are assigned as a putative UDP-glucose dehydrogenase and NAD-dependent epimerase, respectively (31). WbtHFn shows identity to an amidotransferase from F. tularensis subsp. tularensis and may be involved in the formation of the GalNAcAN amido group. Biosynthesis of the remaining sugar, Qui2NAc4NAc, probably requires WbtAFn and WbtN. Sequence similarity suggests that WbtAFn may function as a TDP-glucose 4,6-dehydratase having 99% homology to WbtA (31). WbtN, a predicted aminotransferase, may be involved in Qui2NAc4NAc amination, probably at position 4. Finally, WbtP is a putative acetyltransferase, showing 61% sequence homology to Neisseria meningitidis PglB known to be involved in N-acetyl sugar biosynthesis (30).
TABLE 3.
O-antigen genes and gene products of F. novicida
| Gene or gene product | Length (aa)a | Function | Similar protein(s) | % Identity with F. tularensis SchuS4c | % Positived | Organism | Accession no. |
|---|---|---|---|---|---|---|---|
| IstuF2 | 181b | Transposase | ISFtu2 | 95 (173/181) | 96 (175/181) | F. tularensis | YP_170453 |
| wbtAFn | 368 | dTDP-glucose 4,6-dehydratase | WbtA | 98 (99/101) | 99 (100/101) | F. novicida | YP_170401 |
| wbtN | 371 | Aminotransferase | Unnamed | 64 (235/366) | 80 (293/366) | Bacillus cereus | NP_981686 |
| DegT, DnrJ, EryC1, StrS | 63 (231/365) | 78 (287/365) | Clostridium thermocellum ATCC 27405 | ZP_00510645 | |||
| wbtO | 209 | UDP-galactose phosphate transferase | Unnamed | 62 (128/204) | 80 (165/204) | Oceanobacillus iheyensis | NP_693814 |
| PglC | 62 (126/201) | 76 (154/201) | C. jejuni | ZP_01099808 | |||
| wbtP | 220 | Acetyltransferase | Unnamed | 41 (84/201) | 61 (123/201) | Oceanobacillus iheyensis | NP_693813 |
| PglB | 38 (76/199) | 55 (111/199) | N. meningitidis | AAC25979 | |||
| wbtDFn | 363 | Galacturonosyl transferase | WbtD | 86 (310/359) | 93 (334/359) | F. novicida | YP_170398 |
| wbtEFn | 401 | UDP-glucose/GDP-mannose dehydrogenase | WbtE | 98 (393/401) | 98 (396/401) | F. novicida | YP_170397 |
| wbtFFn | 328 | NAD-dependent epimerase; UDP-glucose 4 epimerase | WbtF | 98 (315/320) | 99 (318/320) | F. novicida | YP_170396 |
| wzy | 430 | Capsule biosynthesis | CapE | 22 (94/427) | 40 (172/427) | Staphylococcus aureus | AAA64644 |
| O-Antigen polymerase | Wzy | 20 (64/314) | 44 (140/314) | E. coli | AAT77177 | ||
| wbtGFn | 371 | Glycosyl transferase | WbtG | 89 (324/361) | 95 (345/361) | F. novicida | YP_170394 |
| wbtQ | 378 | Glycosyl transferase | WbnK | 26 (92/345) | 47 (163/345) | S. dysenteriae | AAO39702 |
| WbqH | Escherichia coli | AAR97066 | |||||
| wbtHFn | 630 | Asparagine synthase; glutamine amidotransferase | WbtH | 97 (611/628) | 98 (618/628) | F. novicida | YP_170393 |
| wzx | 415 | O-Antigen flippase | Wzx | 25 (102/398) | 54 (217/398) | E. coli | AAK60454 |
| Transposase | 156 | Transposase | Transposase | 55 (79/142) | 72 (103/142) | N. meningitidis | AAP44503 |
aa, amino acids.
Partial sequence data.
Values in parentheses indicate number of invariant residues/total number of residues.
Values in parentheses indicate number of invariant and/or conserved residues/total number of residues.
FIG. 1.
Genetic organization of the F. tularensis subsp. tularensis and F. novicida O-antigen gene clusters. (A) F. novicida strain U112. Genes with homologies greater than 90% to those of F. tularensis subsp. tularensis strain SchuS4 have the subscript Fn added to the gene name. Genes unique to F. novicida have no subscript. (B) F. tularensis subsp. tularensis strain SchuS4.
FIG. 2.
Schematic structure of an O-antigen subunit of F. novicida and the putative functions of the gene products of the O-antigen gene cluster. A single unit is shown, with sugar residues and glycosidic linkages indicated (see text).
The gene cluster revealed four potential glycosyltransferases necessary for the biosynthesis of the tetrasaccharide O-antigen repeat unit. One of the first glycosyltransferases is WbtO having identity to PglC of Campylobacter jejuni, which has been demonstrated to add the first sugar residue onto the lipid acceptor undecaprenyl pyrophosphate (13). WbtDFn is another putative glycosyltransferase and possibly transfers GalNAcAN onto Qui2NAc4NAc, based on its homology to WbtD from F. tularensis subsp. tularensis. Elongation to form the GalNAcAN trisaccharide is likely to involve WbtGFn and WbtQ, based on their both showing sequence homology to glycosylatransferases. WbtGFn has homology to WbtG from F. tularensis subsp. tularensis, while WbtQ is proposed to be involved in the formation of GalNAcAN1-4GalNAcAN based on homology to WbnK of Shigella dysenteriae, which also contains the disaccharide as a component of its O-antigen repeat (20). It was noted that WbtP has sequence similarity to N. meningitidis PglB, previously reported to undertake both glycosyltransferase and acetyltransferase functions. It is possible that the WbtP homolog may be bifunctional in F. novicida, although the predicted activity of N. meningitidis PglB glycosyltransferase involves the transfer of a sugar residue onto undecaprenyl pyrophosphate (30), a role similar to that assigned to F. tularensis subsp. tularensis WbtO.
There are two main O-antigen synthesis modes, O-antigen polymerase-dependent and O-antigen polymerase-independent (which can be referred to as wzy dependent and wzy independent) modes. Two genes in the F. novicida O-antigen gene cluster would encode proteins with a high degree of sequence similarity to Wzy and Wzx, suggesting that transportation and polymerization of the O antigen is via a wzy-dependent pathway. Similarly, F. tularensis subsp. tularensis O-antigen biosynthesis is also predicted to proceed via the wzy-dependent pathway (31).
Confirmation of O-antigen structure.
Structural identification of the O antigen of F. novicida strain U112 was achieved using 1D and 2D NMR analysis (results not shown). The 1D spectrum of the lipopolysaccharide O antigen of F. novicida was consistent with the reported structure of F. novicida O-specific polysaccharide (39). It should be noted that the methods used to isolate the O antigen are different from those reported by Vinogradov et al. (39). However, we have also isolated the same material. LPS was isolated from F. novicida and F. tularensis subsp. tularensis with yields of 1.8% and 0.3%, respectively.
Immunization with LPS does not protect against a heterologous challenge.
Immunization with the LPS of strain SchuS4 does not confer full protection against a subsequent challenge with strain SchuS4 (31) but does confer complete protection against a subsequent challenge with F. tularensis subsp. holarctica (12). Therefore, F. tularensis subsp. holarctica was used as the challenge strain in this study.
Mice were immunized with three doses of 50 μg of LPS in 100 μl of phosphate-buffered saline at 14-day intervals. The sera from mice immunized with the LPS of F. novicida had a mean IgM concentration of 4,604 ng/ml and a mean IgG3 concentration of 1,433 ng/ml. In each case these titers were higher than those achieved by immunization with the same level of F. tularensis subsp. tularensis LPS, which were 3,716 ng/ml of IgM antibody and 244 ng/ml of IgG3 antibody. LPS-immunized mice were challenged with 100 MLD of F. tularensis subspecies holarctica or F. novicida.
Mice immunized with LPS from F. novicida were protected against the F. novicida challenge but not against a heterologous F. tularensis subsp. holarctica challenge. Conversely, mice immunized with the F. tularensis subsp. tularensis LPS were not protected from the heterologous F. novicida challenge but were protected from F. tularensis subsp. holarctica challenge (Table 4).
TABLE 4.
Ability for LPS to protect against challenge with F. tularensis or F. novicida
| Immunization | Survival (no. of mice surviving/no. of mice in group) after challenge with:a
|
|
|---|---|---|
| F. tularensis subsp. holarctica | F. novicida | |
| F. tularensis subsp. tularensis LPS | 6/6 | 0/6 |
| F. novicida LPS | 0/6 | 4/6 |
| Naïve | 0/6 | 0/6 |
Survival of mice following immunization with F. tularensis subsp. tularensis or F. novicida LPS and subsequent challenge 21 days postimmunization. No cross-protection was afforded.
Production of defined O-antigen mutants.
Deletion/insertion mutants in genes wbtDEF were made by conjugation of the appropriate modified genes into F. tularensis subsp. tularensis or F. novicida. Southern blotting (Fig. 3) and PCR (results not shown) were used to identify double crossover mutants which were referred to as ΔwbtDEFFt::Cm or ΔwbtDEFFn::Cm for the F. tularensis subsp. tularensis and F. novicida O-antigen mutants, respectively. The deletion/insertion within the O-antigen gene cluster was confirmed by the observation that the probe-reactive band specific to the upstream region was smaller than that of the wild type. This size difference corresponds to 3,052 bp of the wbtDEF region in the SchuS4 sequence compared to only 2,480 bp of the deletion/insertion cassette.
FIG. 3.
Southern blot analysis of DNA from F. tularensis subsp. tularensis and ΔwbtDEFFt::Cm and F. novicida and ΔwbtDEFFn::Cm showing deletion/insertion in the O-antigen clusters. DNA samples were digested with NdeI prior to separation by agarose gel electrophoresis, Southern blotting, and hybridization. The blots were probed with a DIG-labeled probe specific to the upstream flank of wbtDFn/Ft. Lane M, DNA standards of DIG-labeled HindIII-digested λ (molecular size marker II; Roche Diagnostics). (Left) Lane 1, F. tularensis subsp. tularensis SchuS4 DNA; lane 2, ΔwbtDEFFt::Cm DNA with probe-reactive band smaller than that of lane 1. (Right) Lane 1, F. novicida U112 DNA; lane 2, ΔwbtDEFFn::Cm DNA with probe-reactive band smaller than that of lane 1. The faint upper band seen in lane 2 is due to incomplete digestion genomic DNA.
The LPS profiles of ΔwbtDEFFt::Cm and ΔwbtDEFFn::Cm were analyzed by Western blotting using monoclonal antibodies to F. tularensis subsp. tularensis or F. novicida LPS. The monoclonal antibody to F. novicida LPS failed to react with the LPS extract from ΔwbtDEFFn::Cm. Similarly, the monoclonal antibody to F. tularensis subsp. tularensis LPS failed to react with the LPS extract from ΔwbtDEFFt::Cm (Fig. 4).
FIG. 4.
LPS analysis of F. tularensis subsp. tularensis and ΔwbtDEFFt::Cm and F. novicida and ΔwbtDEFFn::Cm. Membranes were probed with monoclonal antibody reactive with F. tularensis subsp. tularensis LPS (A) and monoclonal antibody reactive with F. novicida (B). (A) Lane 1, wild-type F. tularensis subsp. tularensis; lane 2, ΔwbtDEFFt::Cm. The ΔwbtDEFFt::Cm strain has lost all reactivity with the antibody. (B) Lane 1, wild-type F. novicida; lane 2, ΔwbtDEFFn::Cm. The ΔwbtDEFFn::Cm strain has lost all reactivity with the antibody.
The O antigen of F. novicida is critical for serum resistance.
A serum concentration of 0.02% (vol/vol) was sufficient to inhibit 50% of the growth (the MIC50) of the ΔwbtDEFFn::Cm strain. For the wild-type strain U112, the MIC50 was produced with an excess of 50% serum, with the bacteria growing more quickly at the highest concentrations of serum. For the ΔwbtDEFFt::Cm strain the MIC50 of serum was 6.25%, whereas the MIC50 for the wild type was 25% serum.
The O antigen of F. tularensis subsp. tularensis is important for intracellular survival.
To investigate the role of O antigen in macrophage survival and growth, wild-type or O-antigen mutants of F. tularensis were incubated with J774 cells, and the number of intracellular bacteria was determined at intervals. In our macrophage model the ΔwbtDEFFt::Cm and ΔwbtDEFFn::Cm strains behaved differently. Both mutants were taken up at a similar rate to that of their respective wild-type strains. ΔwbtDEFFn::Cm was able to replicate within macrophages. However, although ΔwbtDEFFt::Cm survived in macrophages, it did not replicate over the course of the experiment (Fig. 5).
FIG. 5.
Macrophage uptake and survival assay. The uptake and survival of ΔwbtDEFFt ▪ compared to SchuS4 ○ (A) or of ΔwbtDEFFn ▴ compared to strain U112 ⋄ (B) in J774 macrophages. Intracellular bacteria were counted as the number of CFU per ml of lysed macrophages at 0, 24, and 48 h postinfection. The assays were completed in triplicate with similar results observed each time.
O-antigen mutants of F. tularensis and F. novicida are attenuated but not protective.
The degree of attenuation of the O-antigen mutants of F. tularensis and F. novicida was determined by subcutaneous administration of 105 CFU. For wild-type strains the MLD, by the subcutaneous route, is less than 10 CFU for F. tularensis subsp. tularensis and less than 1,000 CFU for F. novicida (10, 23). Both ΔwbtDEFFt::Cm and ΔwbtDEFFn::Cm were attenuated at the given doses. When ΔwbtDEFFt::Cm-immunized mice were challenged by the subcutaneous route with 100 MLD of SchuS4 49 days postimmunization, there was no protection afforded. Mice immunized with ΔwbtDEFFn::Cm and challenged intraperitoneally with 100 MLD of strain U112 were not protected.
DISCUSSION
Our study has found that F. novicida has an O-antigen biosynthetic gene cluster which contains genes also found in the O-antigen gene cluster of F. tularensis subsp. tularensis strain SchuS4. This is consistent with the similarity of the sugars within the O-antigen repeat subunits. F. novicida is proposed to be the progenitor of F. tularensis subsp. tularensis, F. tularensis subsp. mediasiatica, and F. tularensis subsp. holarctica (38). The evolution of high-virulence strains of F. tularensis may have involved the acquisition of genes and or the rearrangement of the F. novicida O-antigen gene cluster.
Because of the similar compositions of the O antigens from F. tularensis subsp. tularensis and F. novicida, we sought to investigate whether immunization with the LPS from one species could provide heterologous cross-protection. During the intracellular life of F. tularensis and F. novicida, the bacterium escapes the late endosome/phagosome of phagocytic cells to reside in the cytoplasm (reviewed in reference 33). The mechanism by which this happens is unclear. However, it has been proposed that the LPS, or the proposed capsule, is partly shed, and this aids degradation of the phagosomal membrane, permitting the bacterium to “escape” from the phagosome (14). This process could reveal the new or alternative epitopes which may be the same as those exposed by F. novicida. There has previously been a report in the literature of “phase variation” of the F. tularensis subsp. tularensis LPS, with F. novicida-like LPS being expressed (8). This phenomenon could be explained by the shared polysaccharide structures in the O-antigen repeating subunits. Although Cowley et al. have previously reported that some epitopes are common to both LPS types (8), our experiments indicate that at least in mice this is insufficient to afford any cross-protection.
We have also compared the roles of the O antigens of F. tularensis subsp. tularensis and F. novicida in virulence. A previous study by Hartley et al. (17) characterized an O-antigen-deficient (gray) variant of the LVS, but the genetic basis of this phenotypic change is not known, and it is possible that pleiotrophic changes existed. Furthermore, the pathogenesis of disease caused by the attenuated LVS in mice may not fully reflect the pathogenesis of disease caused by virulent strains of F. tularensis and F. novicida. This is the first report of genetically defined O-antigen mutants of F. tularensis and F. novicida.
For some bacteria the O antigen is important for protecting the bacterial cell from complement-mediated killing by serum (1, 24, 25). Our findings indicate that the O antigens of both F. tularensis subsp. tularensis and F. novicida play roles in protection from serum killing. However, the role of the O antigen from F. tularensis subsp. tularensis appears to be relatively minor compared to the role of the F. novicida O antigen. This difference may indicate that in F. tularensis subsp. tularensis other surface structures play a role in resistance to serum killing. In Yersinia pestis surface adhesins (such as the Ail-like proteins) may be involved in serum resistance and compensate for the lack of O antigen (28).
The data presented in this study indicate that, as in many gram-negative bacteria (2, 34, 37), the O antigens of both F. tularensis subsp. tularensis and F. novicida are virulence determinants. However, the functions of the O antigens differ in these two species. Both O-antigen mutants were taken up by macrophages at similar rates to the wild-type strains, but only the ΔwbtDEFFn::Cm strain was able to replicate within the macrophage. The role of the O antigen in promoting the growth of F. tularensis subsp. tularensis in macrophages merits further attention.
The data from this study suggest that the F. tularensis subsp. tularensis O antigen is important for intracellular survival and replication. In contrast, the O antigen of F. novicida is less important in the intracellular niche but more important for the viability of the bacterium before it enters a host cell. Overall, our findings may provide insight into the reasons for the different levels of virulence of F. novicida and F. tularensis subsp. tularensis in humans.
Acknowledgments
We thank Helen Sharps, Sarah Newstead, Donna Ford, and Ronda Griffiths for their technical assistance.
Editor: D. L. Burns
Footnotes
Published ahead of print on 30 October 2006.
REFERENCES
- 1.Anisimov, A. P., S. V. Dentovskaya, G. M. Titareva, I. V. Bakhteeva, R. Z. Shaikhutdinova, S. V. Balakhonov, B. Lindner, N. A. Kocharova, S. N. Senchenkova, O. Holst, G. B. Pier, and Y. A. Knirel. 2005. Intraspecies and temperature-dependent variations in susceptibility of Yersinia pestis to the bactericidal action of serum and to polymyxin B. Infect. Immun. 73:7324-7331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bengoechea, J. A., H. Najdenski, and M. Skurnik. 2004. Lipopolysaccharide O antigen status of Yersinia enterocolitica O:8 is essential for virulence and absence of O antigen affects the expression of other Yersinia virulence factors. Mol. Microbiol. 52:451-469. [DOI] [PubMed] [Google Scholar]
- 3.Bertani, G. 1951. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J. Bacteriol. 62:293-300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chamberlain, R. E. 1965. Evaluation of live tularemia vaccine prepared in a chemically defined medium. Appl. Microbiol. 13:232-235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen, W. X., H. Shen, A. Webb, R. KuoLee, and J. W. Conlan. 2003. Tularemia in BALB/c and C57BL/6 mice vaccinated with Francisella tularensis LVS and challenged intradermally, or by aerosol with virulent isolates of the pathogen: protection varies depending on pathogen virulence, route of exposure, and host genetic background. Vaccine 21:3690-3700. [DOI] [PubMed] [Google Scholar]
- 6.Clackson, T., D. Gussow, and P. T. Jones. 1991. PCR: a practical approach, p. 187-214. In M. J. McPherson, P. Quirke, and G. R. Taylor (ed.), PCR: a practical approach. IRL Press, Oxford, United Kingdom.
- 7.Conlan, J. W., W. X. Chen, H. Shen, A. Webb, and R. KuoLee. 2003. Experimental tularemia in mice challenged by aerosol or intradermally with virulent strains of Francisella tularensis: bacteriologic and histopathologic studies. Microb. Pathog. 34:239-248. [DOI] [PubMed] [Google Scholar]
- 8.Cowley, S. C., S. V. Myltseva, and F. E. Nano. 1996. Phase variation in Francisella tularensis affecting intracellular growth, lipopolysaccharide antigenicity and nitric oxide production. Mol. Microbiol. 20:867-874. [DOI] [PubMed] [Google Scholar]
- 9.Eigelsbach, H. T., and C. Downs. 1961. Prophylactic effectiveness of live and killed tularemia vaccines. I. Production of vaccine and evaluation in the white mouse and guinea pig. J. Immunol. 87:415-425. [PubMed] [Google Scholar]
- 10.Eigelsbach, H. T., and V. G. McGann. 1984. Genus Francisella Dorofe'ev 1947, p. 394-399. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systemic bacteriology, vol. 1. Williams & Wilkins, Baltimore, MD. [Google Scholar]
- 11.Ellis, J., P. C. F. Oyston, M. Green, and R. W. Titball. 2002. Tularemia. Clin. Microbiol. Rev. 15:631-646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fulop, M., P. Mastroeni, M. Green, and R. W. Titball. 2001. Role of antibody to lipopolysaccharide in protection against low- and high-virulence strains of Francisella tularensis. Vaccine 19:4465-4472. [DOI] [PubMed] [Google Scholar]
- 13.Glover, K. J., E. Weerapana, M. M. Chen, and B. Imperiali. 2006. Direct biochemical evidence for the utilization of UDP-bacillosamine by PglC, an essential glycosyl-1-phosphate transferase in the Campylobacter jejuni N-linked glycosylation pathway. Biochemistry 45:5343-5350. [DOI] [PubMed] [Google Scholar]
- 14.Golovliov, I., V. Baranov, Z. Krocova, H. Kovarova, and A. Sjostedt. 2003. An attenuated strain of the facultative intracellular bacterium Francisella tularensis can escape the phagosome of monocytic cells. Infect. Immun. 71:5940-5950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Golovliov, I., A. Sjostedt, A. Mokrievich, and V. Pavlov. 2003. A method for allelic replacement in Francisella tularensis. FEMS Microbiol. Lett. 222:273-280. [DOI] [PubMed] [Google Scholar]
- 16.Gurycova, D. 1998. First isolation of Francisella tularensis subsp. tularensis in Europe. Eur. J. Epidemiol. 14:797-802. [DOI] [PubMed] [Google Scholar]
- 17.Hartley, G., R. Taylor, J. L. Prior, S. Newstead, P. G. Hitchen, H. R. Morris, A. Dell, and R. W. Titball. 2006. Grey variants of the live vaccine strain of Francisella tularensis lack lipopolysaccharide O-antigen, show reduced ability to survive in macrophages and do not induce protective immunity in mice. Vaccine 24:989-996. [DOI] [PubMed] [Google Scholar]
- 18.Hartley, M. G., M. Green, G. Choules, D. Rogers, D. G. C. Rees, S. Newstead, A. Sjostedt, and R. W. Titball. 2004. Protection afforded by heat shock protein 60 from Francisella tularensis is due to copurified lipopolysaccharide. Infect. Immun. 72:4109-4113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hollis, D. G., R. E. Weaver, A. G. Steigerwalt, J. D. Wenger, C. W. Moss, and D. J. Brenner. 1989. Francisella philomiragia comb. Nov. (formerly Yersinia philomiragia) and Francisella tularensis biogroup novicida (formerly Francisella novicida) associated with human disease. J. Clin. Microbiol. 27:1601-1608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Knirel, Y. A., V. V. Dashunin, A. S. Shashkov, N. K. Kochetkov, B. A. Dmitriev, and I. L. Hofman. 1988. Somatic antigens of Shigella: structure of the O-specific polysaccharide chain of the Shigella dysenteriae type 7 lipopolysaccharide. Carbohydr. Res. 179:51-60. [DOI] [PubMed] [Google Scholar]
- 21.Laemmli, U. K. 70 A.D. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. [DOI] [PubMed]
- 22.Larson, C. L., W. Wicht, and W. L. Jellison. 1955. A new organism resembling P. tularensis isolated from water. Public Health Rep. 70:253-257 [PMC free article] [PubMed] [Google Scholar]
- 23.Lofgren, S., A. Tarnvik, G. D. Bloom, and W. Sjoberg. 1983. Phagocytosis and killing of Francisella tularensis by human polymorphonuclear leukocytes. Infect. Immun. 39:715-720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mellata, M., M. Dho-Moulin, C. M. Dozois, R. Curtiss, P. K. Brown, P. Arne, A. Bree, C. Desautels, and J. M. Fairbrother. 2003. Role of virulence factors in resistance of avian pathogenic Escherichia coli to serum and in pathogenicity. Infect. Immun. 71:536-540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Morona, R., C. Daniels, and L. Van Den Bosch. 2003. Genetic modulation of Shigella flexneri 2a lipopolysaccharide O antigen modal chain length reveals that it has been optimized for virulence. Microbiology 149:925-939. [DOI] [PubMed] [Google Scholar]
- 26.Olsufjev, N. G., and I. S. Meshcheryakova. 1982. Infraspecific taxonomy of tularemia agent Francisella tularensis McCoy et Chapin. J. Hyg. Epidemiol. Microbiol. Immunol. 26:291-299. [PubMed] [Google Scholar]
- 27.Olsufjev, N. G., and I. S. Meshcheryakova. 1983. Subspecific taxonomy of Francisella tularensis McCoy and Chapin 1912. Int. J. Syst. Bacteriol. 33:872-874. [Google Scholar]
- 28.Parkhill, J., B. W. Wren, N. R. Thomson, R. W. Titball, M. T. G. Holden, M. B. Prentice, M. Sebaihia, K. D. James, C. Churcher, K. L. Mungall, S. Baker, D. Basham, S. D. Bentley, K. Brooks, A. M. Cerdeno-Tarraga, T. Chillingworth, A. Cronin, R. M. Davies, P. Davis, G. Dougan, T. Feltwell, N. Hamlin, S. Holroyd, K. Jagels, A. V. Karlyshev, S. Leather, S. Moule, P. C. F. Oyston, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001. Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413:523-527. [DOI] [PubMed] [Google Scholar]
- 29.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. [DOI] [PubMed] [Google Scholar]
- 30.Power, P. M., L. F. Roddam, M. Dieckelmann, Y. N. Srikhanta, Y. C. Tan, A. W. Berrington, and M. P. Jennings. 2000. Genetic characterization of pilin glycosylation in Neisseria meningitidis. Microbiology 146:967-979. [DOI] [PubMed] [Google Scholar]
- 31.Prior, J. L., R. G. Prior, P. G. Hitchen, H. Diaper, K. F. Griffin, H. R. Morris, A. Dell, and R. W. Titball. 2003. Characterization of the O antigen gene cluster and structural analysis of the O antigen of Francisella tularensis subsp. tularensis. J. Med. Microbiol. 52:845-851. [DOI] [PubMed] [Google Scholar]
- 32.Sandstrom, G., A. Sjostedt, M. Forsman, N. V. Pavlovich, and B. N. Mishankin. 1992. Characterization and classification of strains of Francisella tularensis isolated in the Central-Asian focus of the Soviet Union and in Japan. J. Clin. Microbiol. 30:172-175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Santic, M., M. Molmeret, K. E. Klose, and Y. bu Kwaik. 2006. Francisella tularensis travels a novel, twisted road within macrophages. Trends Microbiol. 14:37-44. [DOI] [PubMed] [Google Scholar]
- 34.Shankar-Sinha, S., G. A. Valencia, B. K. Janes, J. K. Rosenberg, C. Whitfield, R. A. Bender, T. J. Standiford, and J. G. Younger. 2004. The Klebsiella pneumoniae O antigen contributes to bacteremia and lethality during murine pneumonia. Infect. Immun. 72:1423-1430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shen, H., W. X. Chen, and J. W. Conlan. 2004. Mice sublethally infected with Francisella novicida U112 develop only marginal protective immunity against systemic or aerosol challenge with virulent type A or B strains of F. tularensis. Microb. Pathog. 37:107-110. [DOI] [PubMed] [Google Scholar]
- 36.Simon, R., U. B. Priefer, and A. Pühler. 1983. A broad host range mobilisation system for in vitro genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1:784-791. [Google Scholar]
- 37.Skurnik, M., and J. A. Bengoechea. 2003. The biosynthesis and biological role of lipopolysaccharide O-antigens of pathogenic Yersinia. Carbohydr. Res. 338:2521-2529. [DOI] [PubMed] [Google Scholar]
- 38.Svensson, K., P. Larsson, D. Johansson, M. Bystrom, M. Forsman, and A. Johansson. 2005. Evolution of subspecies of Francisella tularensis. J. Bacteriol. 187:3903-3908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Vinogradov, E., W. J. Conlan, J. S. Gunn, and M. B. Perry. 2004. Characterization of the lipopolysaccharide O antigen of Francisella novicida (U112). Carbohydr. Res. 339:649-654. [DOI] [PubMed] [Google Scholar]
- 40.Vinogradov, E. V., A. S. Shashkov, Y. A. Knirel, N. K. Kochetkov, N. V. Tochtamysheva, S. F. Averin, O. V. Goncharova, and V. S. Khlebnikov. 1991. Structure of the O antigen of Francisella tularensis strain 15. Carbohydr. Res. 214:289-297. [DOI] [PubMed] [Google Scholar]