Previous Article | Next Article ![]()
Infection and Immunity, October 2004, p. 5925-5930, Vol. 72, No. 10
0019-9567/04/$08.00+0 DOI: 10.1128/IAI.72.10.5925-5930.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Institute for Biological Sciences, National Research Council, Ottawa, Ontario, Canada
Received 29 April 2004/ Returned for modification 25 June 2004/ Accepted 1 July 2004
|
|
|---|
|
|
|---|
The chemical structures of the carbohydrate antigens found in the capsules and the O-polysaccharide components (O-PSs) of the lipopolysaccharides (LPSs) of A. pleuropneumoniae serotypes 1 to 12 (27) and serotype 14 (28) have been elucidated and have assisted in explaining observed serological specificities and cross-reactions. Recently, three new serotypes of A. pleuropneumoniae have been described, serotypes 13 and 14, which are NAD independent strains (biovar 2) (26), and serotype 15 (biovar 1), the predominant serotype present in Australian pigs (5).
Knowledge of the serotypes prevalent in a geographic region is important (4) since current vaccines involving killed whole-cell bacterial preparations can protect only against infection by a homologous serotype present in the vaccine (7, 14, 24, 25, 30). We have undertaken the characterization of the newly described serotypes of A. pleuropneumoniae, and this paper describes the elucidation of the structure of the LPS O-PS and the capsular polysaccharide (CPS) of A. pleuropneumoniae serotype 13.
|
|
|---|
Preparation of LPS, CPS, and O-PS. A cell paste of A. pleuropneumoniae serotype 13 (355 g) was extracted with stirred hot 50% aqueous phenol (550 ml, 70°C, 15 min), and the separated phenol and aqueous phases from the cooled (4°C) extract were dialyzed against tap water until phenol free and then lyophilized. The products were resuspended in 60 ml of 0.02 M sodium acetate (pH 7.0) and were treated sequentially with RNase, DNase, and proteinase K (37°C, 3 h each). Trace solids were removed from the digest by low-speed centrifugation, the samples were subjected to ultracentrifugation (105,000 x g, 4°C, 12 h), and the precipitated gels were dissolved in water and lyophilized to yield 2.22 g (aqueous phase) and 84 mg (phenol phase) of LPS.
The
supernatants were treated with cold acetone (6 volumes), and the
precipitated products (crude CPS) were collected by centrifugation. The
aqueous-phase precipitate (1.90 g) of crude putative CPS upon Sephadex
G-50 gel filtration yielded a high-molecular-mass polysaccharide
(Kav, 0.03 to 0.04; 320 mg
[distribution coefficient Kav =
(Ve V0)/(Vt
V0), where Ve is the elution volume of the
specific material, V0 is the void volume of the system, and
Vt is the total volume of the system]) with an
[
]D + 99o (c 0.3,
water) that was used in subsequent
analyses.
O-deacetylated CPS was prepared by
treatment of native CPS (30 mg) with 0.02 M ammonium hydroxide (5 ml)
at 37°C for 4 h and following dilution with water was
lyophilized to yield acetate-free CPS (27 mg) with an
[
]D + 62o (c 1.4,
water).
Aqueous-phase LPS (0.5 g) was hydrolyzed with 2% (vol/vol) acetic acid (AcOH) (75 ml, 100°C, 2 h), and following the removal of precipitated lipid A (117 mg) the lyophilized water-soluble products were fractionated by Sephadex G-50 column chromatography to yield a high-molecular-mass O-PS (Kav, 0.03 to 0.04; 224 mg), a core oligosaccharide (Kav, 0.34; 80 mg), and a low-molecular-mass fraction (Kav, 0.94; 40 mg) containing KDO (3-deoxy-D-octulosonic acid).
Chromatography. Gas chromatography was performed using a ZB-50 column (30 m by 0.25 mm; Phenomenex) in an Agilent 6850 chromatograph fitted with a flame ionization detector or a Varian Saturn 2000 ion-trap gas chromatography-mass spectrometry (MS) instrument and a temperature program of 170°C (delay, 2 min) at 2°C/min to 220°C. Retention times and mass spectra were matched with authentic reference samples. Gel filtration chromatography was done with either Sephadex G-50 or Bio-gel P2 columns as previously described (20).
NMR spectrometry. For nuclear magnetic resonance (NMR) spectrometry, 1H, 13C, and 31P spectra were recorded with a Varian 400-MHz spectrometer with samples in D2O and referenced to an internal acetone standard (1H, 2.225 ppm; 13C, 31.07 ppm). COSY, TOCSY, NOESY, HSQC, and HMBC experiments were done as previously described (31).
Periodate oxidation. Smith-type oxidation (11) of O-PS was done as previously described (1).
O deacetylation. CPS (29 mg) dissolved in 0.02 M ammonium hydroxide (5 ml) was kept at 37°C for 4 h, and following dilution with water (10 ml) and lyophilization, the void volume fraction obtained by Sephadex G-50 gel filtration was collected, lyophilized (yield, 27 mg), and used for NMR analysis.
Colorimetric analyses. Determination of O-acetyl was made as described by Hestrin (15), and phosphate was determined by the method of Chen et al. (6).
|
|
|---|
Mild acid hydrolysis of the LPS (2% AcOH, 100°C, 2 h) yielded a precipitate of released lipid A (23%), and gel filtration column chromatography of the water-soluble products gave a high-molecular-mass O-PS (45%), a core oligosaccharide fraction (16%), and a low-molecular-mass fraction containing KDO (8%).
The O-PS
had an [
]D + 24o (c
0.3, water) and on acid hydrolysis yielded L-rhamnose,
D-galactose, and 2-amino-2-deoxy-D-galactose in
the molar ratio 1:2:1. Preparative paper chromatography afforded
chromatographically pure samples of each glycose, which were fully
characterized from their paper chromatographic mobilities, gas-liquid
chromatography (GLC)-MS analysis of their reduced (sodium borodeuteride
[NaBD4]) and acetylated alditol-1d
derivatives (12), and
determination of their configurations by GLC analysis of their
acetylated derived 2-(R)-butyl-glycosides
(10).
Consideration of the glycose composition and the one-dimensional (1D) 1H and 13C NMR spectra of the O-PS indicate that the O antigen is composed of a repeating tetrasaccharide unit. The proton spectrum (Fig. 1; Table 1) showed four equal-area anomeric proton resonances (4.42 to 5.03 ppm), together with signals at 1.30 (3H) ppm arising from H-6 of the L-Rha constituent and at 2.03 ppm (3H) from the N-acetyl (CH3) substituent of the 2-acetamido-2-deoxy-D-galactose (D-GalNAc) constituent. Consistent with the proton NMR data, the 13C NMR spectrum (Table 1) showed four anomeric carbon signals (102.0 to 105.4 ppm), a signal at 17.5 ppm from the C-6 of the L-Rha constituent, a resonance at 52.6 ppm characteristic for a 2-deoxy-2-acetamido C-2 substitution of a D-GalNAc residue, and inter alia resonances at 22.8 ppm (NH-COCH3) and 175.5 ppm (NH-COCH3), also characteristic of the N-acetyl substitution of the GalNAc O-PS constituent.
![]() View larger version (12K): [in a new window] |
FIG. 1. 1H
NMR spectrum of the O-PS of A. pleuropneumoniae
serotype 13. Indicated are the four anomeric signals (H-1) of the
component glycose units:
4)- -L-Rha-(1 (A),
3)-ß-D-Gal-(1 (B),
3,4)-ß-D-GalNAc-(1 (C), and
ß-D-Gal-1 . Also indicated are the NAc
methyl signal of C and the C-6 methyl signal of
A.
|
|
View this table: [in a new window] |
TABLE 1. NMR
spectral data for the LPS O polysaccharide of A.
pleuropneumoniae serotype 13
|
4)-L-Rhap-(1
,
D-Galp-(1
,
3)-D-Galp-(1
and
3,4)-D-GalNAc(1
.
2D NMR methods were
employed to determine the interglycosidic linkages and the anomeric
configurations of the above units within the O-PS. The assignments of
chemical shifts of the proton and carbon atoms in the O-PS were
established from COSY and TOCSY analysis, and carbon-proton
correlations were established from HSQC and HMBC measurements (Fig.
2; Table
1). From the respective coupling constants JH-1,H-2
(<2 Hz) and JC-1,H-1 (171 Hz), the
L-Rhap could be assigned the
configuration, while D-Galp and
D-GalpNAc, with corresponding coupling constants of
7.2 to 8.2 Hz and 163 to 165 Hz, respectively, allowed them to be
assigned the ß configuration. A 2D phase-sensitive NOESY
experiment provided interresidue data consistent with the methylation
linkage analysis, thus establishing the structure of the O-PS
as:
![]() |
![]() View larger version (13K): [in a new window] |
FIG. 2. HSQC
NMR spectrum of the O-PS of A. pleuropneumoniae serotype 13
showing proton and carbon correlation cross peaks for the component
glycose units: 4)- -L-Rha-(1 (A),
3)-ß-D-Gal-(1 (B),
3,4-ß-D-GalNAc-(1 (C), and
ß-D-Gal-(1
(D).
|
![]() |
The accumulated experimental data provide convincing evidence that the structure of the LPS O-PS of A. pleuropneumoniae serotype 13 is identical to that previously determined for the O antigen of A. pleuropneumoniae serotype 7 (3). The O-PS structure is similar to that of the O-PS of A. pleuropneumoniae serotype 4 (2), differing only in the replacement of the D-GalpNAc by a 1,3-linked ß-D-Glcp side-group residue instead of a 1,3-linked ß-D-Galp side-group residue. It is interesting that Lebrun et al. (19) showed that monoclonal antibodies can be specific or nonspecific for the respective O-PSs of A. pleuropneumoniae serotypes 4 and 7, a finding consistent with the chemical structures, which suggest that the O-PSs should share similar carbohydrate epitopes and yet should also have unique structural features involving single nonreducing ß-D-Galp or ß-D-Glcp side end-group residues.
In order to prevent serological misidentification of A. pleuropneumoniae serotype 13 due to cross-reacting antibody to common LPS O antigen, specific type 13 antibody may be made by using A. pleuropneumoniae serotype 13 CPS or its conjugate as an immunogen. Alternatively, polyclonal antisera prepared against whole A. pleuropneumoniae serotype 13 cells may be adsorbed out by A. pleuropneumoniae serotype 7 cells or by insolublized LPS (8) from either A. pleuropneumoniae serotype 7 or 13.
The water phase of the
phenol-extracted A. pleuropneumoniae serotype 13 cells, after
removal of LPS by ultracentrifugation and following precipitation with
acetone, yielded a high-molecular-mass product which upon purification
by Sephadex G-50 column chromatography gave a fraction
(Kav, 0.01 to 0.05) which had an
[
]D +99o (c 0.3,
water) and on hydrolysis yielded D-galactose and glycerol
(1:1), identified by GLC-MS of derived glycitol acetates, the
D-Gal being further characterized by GLC of its acetylated
2-(R)-butyl glycoside derivatives. Colorimetric phosphate
analysis (6) showed the
product to contain 7.1% phosphorus, leading to the conclusion
that the capsular material was composed of equimolecular amounts of
D-galactose, glycerol, and phosphate. Colorimetric analysis
(15) indicated that the
native CPS contains a total 1.0 M O-acetyl substitution of the
D-Galp residue.
Mild basic hydrolysis or
dephosphorylation of the CPS with cold 48% aqueous HF resulted
in depolymerization products and the quantitative chromatographic
isolation of compound III, composed of D-galactose and
glycerol (1:1), and it had an [
]D
+ 158o (c 0.2, water). The 1H and
13C NMR spectra of compound III showed chemical shifts (Fig.
3; Table
2) that identified it as
-D-galactopyranosyl-(1
2)-glycerol. The
spectra showed all the expected signals for an
-D-Galp linked to the O-2 of a glycerol
residue, following from the observation of an NOE correlation between
Gal H-1 and glycerol (Gro) H-2. The glycerol being phosphorylated at
O-1 and O-3 led to low-field shifts of H-1, H-3, C-1, and C-3, as seen
in comparing 1H and 13C to their shifts in
O-deacetylated CPS (Table
2). The occurrence of the anomeric carbon signal of the
D-Galp residue seen at 97.5 ppm with J
C-1,H-1 174 Hz with the corresponding proton signal at 5.15
ppm with a coupling constant JH-1,H-2 3.8 Hz,
considered in conjunction with the high positive specific optical
rotation of compound III, confirms the
configuration of the
D-Galp moiety and the proposed structure. The
characterization of compound III indicates that the CPS is a polymer of
a basic backbone chain of C-1 and C-3 phosphate ester-linked glycerol
residues bearing single
-D-Galp residues
glycosidically linked at the C-2 positions of each glycerol
unit.
![]() View larger version (9K): [in a new window] |
FIG. 3. HSQC
NMR spectrum of compound III from A. pleuropneumoniae serotype
13 CPS showing carbon and proton correlation cross peaks for the
-D-galactose (Gal) and glycerol (Gro)
components.
|
|
View this table: [in a new window] |
TABLE 2. NMR
spectral data for compound III from the CPS and the O-deacetylated CPS
of A. pleuropneumoniae serotype 13
|
-D-Gal O-3 and O-4 positions compared with the
chemical shifts observed for the same proton and carbon atoms in the
O-deacetylated CPS. While it is known that the presence of O-acetyl
groups can determine antigenic specificities
(18), it is also known
that the degree and position of O acetylation are often dependent on
cell culture conditions, and batch-to-batch production changes in
O-acetyl position and degree of substitution are often found. In the
present study some batch CPS preparations showed a minor
(<4%) O-acetyl substitution at the C-2 Galp
position. The combined results of the chemical and physical analyses of
the currently studied A. pleuropneumoniae serotype 13 CPS show
the polymer to have structure IV.
![]() |
The structural occurrence of 1,3 poly(glycerol phosphates) (type I teichoic acids) is not unusual. They have been found in many bacterial species (21, 22, 29), and glycosyl substitution at C-2 by the monosaccharide units of glucose, galactose, rhamnose, and 2-acetamido-2-deoxyhexose have been demonstrated.
We thank the Canadian Bacterial Diseases Network, Center of Excellence, for their support.
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»