Previous Article | Next Article 
Journal of Bacteriology, January 1999, p. 656-661, Vol. 181, No. 2
0021-9193/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.
Role of Glycosylation at Ser63 in Production of
Soluble Pilin in Pathogenic Neisseria
Michael
Marceau1,2 and
Xavier
Nassif1,*
INSERM U411, Laboratoire de Microbiologie,
Faculté de Médecine Necker-Enfants Malades, 75015 Paris,1 and
Laboratoire de
Bactériologie, Faculté de Médecine Henri
Warembourg, 59045 Lille,2 France
Received 8 September 1998/Accepted 30 October 1998
 |
ABSTRACT |
Pilus-mediated adhesion is essential in the pathogenesis of
Neisseria meningitidis (MC) and Neisseria
gonorrhoeae (GC). Pili are assembled from a protein subunit
called pilin. Pilin is a glycoprotein, and pilin antigenic variation
has been shown to be responsible for intrastrain variability with
respect to the degree of adhesion in both MC and GC. In MC,
high-adhesion pilins are responsible for the formation of bundles of
pili which bind bacteria and cause them to grow as colonies on infected
monolayers. In this work, we selected MC and GC pilin variants
responsible for high and low adhesiveness and introduced them into the
other species. Our results demonstrated that a given pilin variant
expressed an identical phenotype in either GC or MC with respect to
bundling and adhesiveness to epithelial cells. However, the production of truncated soluble pilin (S pilin) was consistently more abundant in
GC than in MC. In the latter species, the glycosylation of pilin at
Ser63 was shown to be required for the production of a truncated
monomer of S pilin. In order to determine whether the same was true for
GC, we engineered various pilin derivatives with an altered Ser63
glycosylation site. The results of these experiments demonstrated that
the production of S pilin in GC was indeed more abundant when pilin was
posttranslationally modified at Ser63. However, nonglycosylated
variants remained capable of producing large amounts of S pilin. These
data demonstrated that for GC, unlike for MC, glycosylation at Ser63 is
not required for S-pilin production, suggesting that the mechanisms
leading to the production of S pilin in GC and MC are different.
| |
INTRODUCTION |
Neisseria gonorrhoeae
(GC) and Neisseria meningitidis (MC) are pathogens belonging
to the same genospecies (4). They both have type IV pili,
which are essential in virulence, promoting bacterial interactions with
eucaryotic cells (15, 28). Pili are filamentous structures
on bacterial surfaces and are assembled from a protein subunit
designated PilE or pilin. Besides that of pilin, the expression of one
of the PilC proteins is required for piliation (8).
There are two PilC proteins, designated PilC1 and PilC2. These proteins
are highly homologous and are found in both pili and outer membranes
(8, 21, 22). PilC proteins play an important role in
bacterial adhesiveness and are considered to be tip-located adhesions
(22). In GC, these proteins have similar functions in
piliation and adhesion, and PilC1+
PilC2
and PilC1 PilC2+ strains
have similar phenotypes. In MC, only PilC1 is capable of
promoting adhesion and piliation, whereas PilC2+
PilC1 strains are piliated but are incapable of
pilus-mediated adhesion. The reasons why PilC1
PilC2+ MC isolates are nonadhesive remain unknown.
MC produces two types of pilin. Class I pilins are recognized by their
ability to bind monoclonal antibody SM1, which reacts with an epitope
localized in the constant region, as described by Virji et al.
(29), while class II pilins are not. MC class I pilins show
extensive homology with GC pilins (20). In both species,
pilin undergoes antigenic variation (6, 18), and pilin
antigenic variation has been shown to modulate bacterial adhesiveness
(11, 16, 23, 30). In MC, high-adhesion pilin variants are
responsible for the formation of large bundles of pili (13)
which bind bacteria and cause them to grow as colonies on infected
monolayers. On the other hand, strains expressing low-adhesion variants
have long flexible pili which do not form bundles, and bacteria grow as
isolated diplococci on the surface of cell monolayers. In GC, the
mechanism by which pilin antigenic variation modulates bacterial
adhesiveness has not yet been explored.
As for all type IV pilins, the product of the pilE gene is a
precursor that is processed at a highly conserved consensus cleavage site, located close to the N terminus. This cleavage, which removes 7 amino acids, requires the product of the pilD gene, a
prepilin peptidase. Some pilin variants are processed at an additional cleavage site, at which 39 amino acids are removed from the N terminus.
These latter truncated forms of pilin, designated soluble pilin (S
pilin), are not assembled into pili and are secreted into the
surrounding media. Strains expressing such variants are poorly piliated
and subsequently poorly adhesive (5).
Recently, MC and GC pili were found to be glycosylated. Atomic
resolution of the structure of a GC pilin revealed that
galactose-
-1,3-N-acetylglucosamine was O linked to Ser63
(19). On the other hand, Stimson et al. (26)
showed that an O-linked trisaccharide was present between amino acid
residues 45 and 73 of MC pilin. This structure contains a terminal
1,4-linked digalactose covalently linked to a
2,4-diacetamido-2,4,5-trideoxyhexose (26). Recently, by use
of a class I pilin MC strain, evidence that
galactose--1,3-N-acetylglucosamine was O linked to Ser63 was obtained. This finding demonstrates that the difference in glycosylation patterns is not an intrinsic difference between MC and
GC, since a strain-to-strain difference exists among MC strains. By
engineering pilin variants with altered Ser63 glycosylation sites, we
demonstrated that glycosylation is required for the production of S
pilin in MC (14). However, a similar role for glycosylation
in GC has not been reported.
In this work, pilin variants responsible for high and low adhesiveness
and initially obtained from MC were introduced into GC and vice versa.
Our data demonstrated that a defined pilin variant was responsible for
the same pilus morphology and the same adhesive phenotype in both MC
and GC. However, the amount of S pilin produced by the variant was
always more abundant in GC than in MC. Construction of various
derivatives with modified Ser63 glycosylation sites confirmed that a
posttranslational modification at Ser63 increased the production of S
pilin in both GC and MC. However, nonglycosylated variants remained
capable of producing S pilin in GC but not in MC. Taken together, these
data demonstrate that the mechanisms which lead to the production of S
pilin are different in these pathogenic neisseriae.
| |
MATERIALS AND METHODS |
Bacterial strains, growth conditions, and oligonucleotides.
Strain 8013 is an encapsulated serogroup C class I MC strain
(17). FA1090 and MS11 are previously described GC strains
(2, 27). Variants of these strains are shown Table
1. The SA(Ser62Ala-Ser63Ala) pilin has
been previously described. This variant, in which Ser62 and Ser63 of SA
are replaced by two Ala residues, is nonglycosylated (14).
GC and MC strains were routinely grown on GCB (Difco) agar containing
the supplements described by Kellog et al. (9). All experiments performed throughout this work were done with overnight cultures from frozen stocks. Strains 8013 and FA1090 were transformed by previously described techniques (16). Kanamycin was used at a concentration of 100 µg/ml for selection of both MC and GC.
The sequences of oligonucleotides PILEM1, PILEM2, and KM5 were
5'-CCCTTATCGAGCTGATGATTG-3', 5'-CAGCCAAAACGGACGACCCC-3',
and
5'-GGAGACATTCCTTCCGTATC-3', respectively. The
complementary oligonucleotides
11GAS+
(5'-CCCGCCGACAACAGTTCTGCCGGCGTGGCA-3') and 11GAS
were
designed in order to replace Gly62 and Ala63 of RM11 pilin with
two Ser
residues and correspond to the coding and noncoding strands,
respectively. The sequences of oligonucleotides PILEM3ECO and
PILEM2ECO are 5'-GCGAATTCACCGACCCAATCAACACACCCG-3' and
5'-CGGAATTCAGCCAAAACGGACGACCCC-3',
respectively. The
locations of these oligonucleotides on the
pilE gene are
shown in Fig.
1.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 1.
Organization of the pilE locus of GC and
class I MC strains. Residues 1 to 53 of the mature pilin correspond to
the constant region, and residues 54 to 161 correspond to the variable
and hypervariable (hv) regions. Black boxes indicate the locations of
the conserved regions. pilE::km fusion constructs
were obtained as previously described (13, 14, 16).
Oligonucleotide PILEM3ECO anneals to a segment located upstream of the
pilE promoter (region a). Oligonucleotide PILEM1 is in
region b. Oligonucleotides PILEM2 and PILEM2ECO are in region e, and
KM5 is in region d. The mutagenizing oligonucleotides (11GAS+ and
11GAS) are in region c. The detail of the sequence between residues
53 and 77 demonstrated the existence of a Ser at positions 62 and 63 for SA, SB, and SC and at position 63 for 308. Ser63 is known to be O
linked to a carbohydrate residue in both species. No Ser63 was present
in the RM11 pilin variant. The detail of the sequence of the C-terminal
region (residues 118 to 161) showed that at position 140, there is a
Lys in variants responsible for bundling and high adhesion (SB and 308)
and a Glu in variants responsible for low adhesion (SA, SC, and RM11)
(13). Asterisks indicate conserved residues. The figure is
not drawn to scale.
|
|
Construction of pilE::Km transcriptional
fusions and site-directed mutagenesis.
Standard molecular biology
techniques were performed as described by Sambrook et al.
(24). The construction of the pilE::Km fusions has already been described (16). They contained the following fragments, from 5' to 3': (i) the pilin open reading frame,
(ii) the fragment carrying the aph3' kanamycin resistance gene lacking its own promoter, (iii) the neisserial DNA uptake sequence, and (iv) a 120-bp sequence corresponding to the fragment located downstream of the meningococcal pilE stop codon.
PilE mutants of GC and MC were obtained by transforming a
construction containing a truncated pilin open reading frame fused to
aph3'.
The
pilE::Km fusions were introduced by
transformation into 8013 (clone 12) and the derivative of FA1090
expressing the RM11
pilin variant. Km
r transformants
arising by recombination with the
pilE locus were
selected,
and transformants expressing the appropriate variants
were chosen by
sequencing of the
pilE locus. Chromosomal DNA of
these
transformants was retransformed into FA1090 and 8013 (clone
12). Pools
of about 1,000 Km
r colonies were then frozen. All
biological assays were performed
with these pools in order to eliminate
the possibility of phenotypic
change due to variation of a bacterial
component other than
pilin.
For expression in
E. coli, the pilin from the above
transformants was first amplified between PILEM3ECO and PILEM2ECO and
cloned into the
EcoRI site of pBR325 before being introduced
into
strain
MC1061.
Site-directed mutagenesis of RM11 pilin was performed by PCR overlap
extension (
7) with oligonucleotides PILEM1 and PILEM2
and a
set of two mutagenizing oligonucleotides (11GAS+ and 11GAS
).
Mutagenesis was performed as previously described (
13,
14).
The template was total DNA of clone 12 expressing the
RM11::Km
variant.
Protein preparation, antibodies, and immunoblots.
Whole-cell
lysates were denatured and electrophoresed by sodium dodecyl sulfate
(SDS)-polyacrylamide gel electrophoresis (PAGE) by standard protocols
(10). Monoclonal antibody 5C5 recognizes a conserved epitope
in the constant region of class I pilin (14). Polyclonal
anti-PilC antiserum was a gift from A.-B. Jonsson (15). Monoclonal antibody 4B12, recognizing Opa proteins, was a gift from M. Blake (1).
Cell culture and adherence assays.
Assays were performed a
minimum of three times for each strain with Hec-1B epithelial cells as
previously described (16). Bacterial overnight cultures were
resuspended in cell culture media at a density of 106 CFU
ml1. Wells containing Hec-1B cells were incubated at
37°C with 1 ml of bacterial suspension under 5% CO2.
After 4 h of incubation, the number of CFU present in the
supernatant was determined. Each well was then washed three times, the
mammalian cells were lifted off the plates, and the number of bacteria
associated with the mammalian cells was calculated by serial dilution.
Adhesiveness was determined as the ratio of cell-associated CFU to CFU
in the supernatant. Low-adhesion strains were defined as having a ratio of 
0.05. The adhesiveness of strains expressing high-adhesion variants was at least four to five times higher than that of strains expressing low-adhesion variants.
Monitoring of piliation status by electron microscopy.
Bacteria were grown overnight on GCB agar plates without antibiotics,
resuspended in phosphate-buffered saline at a density of
106 CFU ml1, adsorbed to carbon
Formvar-coated nickel grids, and washed several times with water. After
being stained with 2% uranyl acetate for 1 min, the samples were
viewed and photographed.
Quantification of pilin amounts by density analysis.
In
order to appreciate the respective amounts of S pilin and full-length
monomer produced by each variant, Western blot membranes were scanned,
and one-dimensional densitometry analysis was performed on a Macintosh
G3 computer with the public-domain NIH Image application (http://rsb.info.nih.gov/nih-image/). The ratio, designated
RS, corresponding to S pilin divided by total
pilin (i.e., S pilin plus full-length pilin) was determined. For each
variant, this ratio was obtained from at least four different
experiments performed with various protein concentrations.
| |
RESULTS |
Interspecies exchanges of the pilE gene.
In a
first set of experiments, defined pilin variants isolated from MC or GC
and responsible for high and low adhesiveness were introduced into the
species that they had not originated from. SB, SA, and SC pilins were
obtained from variants of MC strain 8013 (Table 1), an encapsulated
serogroup C class I strain. In 8013, SB pilin is responsible for high
adhesiveness, whereas SA and SC pilins are associated with low
adhesiveness. MC isolates producing the SA variant have been shown to
release S pilin in addition to full-length mature pilin
(14). In previous work, we demonstrated that high-adhesion
MC pilin variants, which are responsible for the formation of bundles
of pili, have a Lys at position 140, whereas low-adhesion MC pilin
variants have a negatively charged amino acid in this location
(13) (Fig. 1).
RM11 and 308 GC pilin variants were initially obtained from GC strains
FA1090 and MS11, respectively (yielding strains FA1090-RM11
and
MS11-308, respectively) (Table
1). The RM11 PilE variant
produces
S pilin. The adhesiveness of the wild-type strains expressing
these
variants is shown in Table
1 and is compared to that of
a nonpiliated
derivative of strain FA1090. The nonpiliated GC
adhered much better
than the nonpiliated variant MC, probably
as a consequence of both MC
encapsulation and expression of Opa
proteins by FA1090. The
adhesiveness of RM11 and 308 relative
to that of FA1090
pilE::Km was compatible with low- and
high-adhesion
phenotypes, respectively. Furthermore, the pilus
morphology of
these strains was consistent with these data, since
308 is responsible
for the formation of bundles of pili, whereas RM11
produces long
and flexible pili without any bundling. However, it
should be
pointed out that the piliation of the latter strain was poor,
probably due to the large amount of S-pilin production. The primary
sequences of 308 and RM11 pilins contained a Lys and a Glu,
respectively,
at position 140, as expected for high- and low-adhesion
variants,
consistent with previous data obtained with MC pilin variants
(Fig.
1).
The
pilE gene corresponding to each of the above variants
was transcriptionally fused to a Km
r gene lacking its own
promoter. These fusions were then introduced
by allelic
replacement into the
pilE locus of MC clone 12 and
GC
strain FA1090. The former is an Opa
PilC
+ derivative of MC strain 8013, and the latter is
Opa
+ PilC
+. In order to eliminate the
possibility of phenotypic changes
due to the variation of a
bacterial component other than pilin,
all experiments described below
were carried out with pools of
transformants produced as described in
Materials and
Methods.
In a first set of experiments, the electrophoretic properties of each
variant, when produced in GC or MC, were compared. The
amount of S
pilin produced was also determined with
RS.
The results
are reported Fig.
2.
Pilins of strains expressing SB::Km, SC::Km,
and 308::Km fusions were processed in an identical manner in
both
Neisseria species. On the other hand, differences were
observed
with S-pilin-producing strains, i.e., SA::Km and
RM11::Km variants.
The SA::Km fusion was
responsible for the production of a larger
amount of truncated S pilin
in GC than in MC (Fig.
2, lanes 4G
and 4M). The RM11::Km
fusion was responsible for the production
of S pilin only in GC (Fig.
2, lane 5G) (
RS, 0.54 ± 0.11); in
MC, only
full-length monomers were detected (Fig.
2, lane 5M)
(
RS, 0 ± 0). These data suggested that the
mechanisms involved
in S-pilin production are not identical in MC and
GC.
View larger version (37K):
[in this window]
[in a new window]
|
FIG. 2.
Western blot of total bacterial extracts showing the
electrophoretic migration of variants containing the SB::Km
(lanes 1), SC::Km (lanes 2), 308::Km (lanes 3),
SA::Km (lanes 4), and RM11::Km (lanes 5) pilin
fusions. Each variant was produced by MC strain 8013 (lanes M) or GC
strain FA1090 (lanes G). The values below the lanes correspond to the
mean (standard deviation) RS values determined
by density analysis in at least four independent experiments containing
various quantities of proteins. Migration was identical for each
variant in both species, except for S-pilin-producing variants (SA and
RM11), which produced larger amounts of truncated S pilin in GC than in
MC.
|
|
Adhesiveness induced in both species by each
pilE::Km fusion was next determined (Table
2). SB::Km- and
308::Km-expressing
strains were at least fivefold more
adhesive than isogenic strains
expressing low-adhesion pilin fusions
(SC::Km, SA::Km, and RM11::Km)
in both MC
and GC, thus demonstrating that pilin variants are
responsible for
similar adhesion phenotypes in both species. However,
it should be
pointed out that adhesiveness induced by these fusions
was always lower
than that of the wild-type strain expressing
pilE alleles
(compare values in Table
2 and Table
1). The likely
explanation is a
modification of mRNA stability due to transcriptional
fusion of the
pilE gene with the Km
r gene.
Comparative piliation analysis of the above transformants was carried
out by transmission electron microscopy (Fig.
3). In
both species, high-adhesion SB and
308 pili formed bundles, whereas
low-adhesion SC and SA pili formed
long, separate fibers (
13).
In isolates expressing the
RM11::Km fusion, long, separate pili
were detected only in
MC; in GC, no individual fibers were clearly
visualized by transmission
electron microscopy (data not shown).
This latter difference is
consistent with the fact that RM11 produced
a larger amount of S pilin
in GC than in MC. However, it should
be pointed out that the
wild-type strain FA1090 RM11 is piliated,
albeit rather
poorly. Furthermore, FA1090 expressing the RM11::Km
fusion was more adhesive than a nonpiliated derivative (Table
2). Taken
together, these data suggest that a smaller number
of pili might be
produced by FA1090 expressing RM11::Km than by
the wild-type
strain FA1090-RM11 and not detected by electron
microscopy. As
mentioned above, the reduction in pilin production
due to
transcriptional fusion might be responsible for the differences
in
piliation observed between FA1090-RM11 and FA1090 expressing
RM11::Km.
View larger version (41K):
[in this window]
[in a new window]
|
FIG. 3.
Transmission electron microscopy of negatively stained
pilin preparations of MC or GC transformants containing the
SB::Km, SC::Km, 308::Km,
SA::Km, and RM11::Km pilin fusions. Pili were
produced by the species from which the original variant was not
isolated; hence, SB, SC, and SA pili are shown as being produced in GC
strain FA1090, and 308 and RM11 pili are shown as being produced in MC
strain 8013. SB and 308 pili make bundles, unlike SC, SA, and RM11
pili.
|
|
Taken together, the above data demonstrate that a given pilin variant
expressed nearly identical phenotypes in both GC and
MC with respect to
bundling and adhesiveness. The only difference
observed was in the
amount of S pilin produced, which was always
larger in GC than in
MC.
Consequences of pilin glycosylation on Ser63 for the production of
S pilin.
One striking finding from the above data was that, in
strains expressing RM11 and SA, the amount of S pilin produced was
larger in GC than in MC. Furthermore, in a previous study, we showed that O-linked glycosylation on Ser63 of an SA variant was required for
MC to enhance pilin production and that, in variants altered at this
site, the release of truncated S pilin was dramatically reduced.
Consistent with these data is the lack of S-pilin production in MC by
the RM11 variant, which does not have a Ser at position 63 (Fig. 1).
However, strikingly significant amounts of S pilin were produced by
this variant in GC. These results suggest that, in these two species,
the contributions of glycosylation to S-pilin production could be different.
In order to investigate the possible role of a posttranslational
modification on Ser63 in the maturation of RM11 GC pilin
into S pilin,
site-directed mutagenesis of RM11 was performed.
The RM11 variant
contains a Gly at position 62 and an Ala at position
63 (Fig.
1). A
derivative of RM11 was engineered to contain Ser
at positions 62 and
63. The resulting pilin variant was designated
RM11rg. In
Neisseria, the molecular weight of the full-length
monomeric
pilin of RM11rg was higher than that of the parental
RM11 pilin
(Fig.
4B, compare lanes 1M and 1G with
lanes 2M and
2G), demonstrating that the replacement of an Ala by a Ser
restored
a posttranslational modification which does not occur in
Escherichia coli. The molecular weights of RM11rg and RM11
pilins produced
in
Neisseria were next compared to those of
these pilins produced
in
E. coli (Fig.
4B). For this latter
species, it has clearly
been demonstrated that, due to the lack of
prepilin peptidase,
incomplete processing occurs and gives rise to two
different products
of the pilin gene: a large one, prepilin, whose
signal peptide
is not cleaved, and a smaller one (790 Da), which
corresponds
to mature pilin (
3). As expected, two such bands
were seen
when RM11 and RM11rg were produced in
E. coli
(Fig.
4, lanes 1E
and 2E). Furthermore, the migration of these two
variants produced
in
E. coli was identical, showing that
changing an Ala to a Ser
did not intrinsically affect the protein
structure; this result,
however, does not explain the modification of
migration observed
between Fig.
4B, lanes 2M and 2G, on the one hand
and lanes 1M
and 1G on the other hand. These findings confirm the
introduction
of a posttranslational modification in RM11.
View larger version (36K):
[in this window]
[in a new window]
|
FIG. 4.
Western blot of total bacterial extracts. (A) Samples
were electrophoresed for 30 min in a gel containing SDS, a 13 to 20%
acrylamide gradient, and 4 M urea. (B) Same gel but electrophoresed for
90 min. Lanes 1, nonglycosylated RM11 pilin variant produced by MC (M),
GC (G), or E. coli (E). Lanes 2, glycosylated RM11rg pilin
variant produced by MC (M), GC (G), or E. coli (E).
Monoclonal antibody 5C5 was used for pilin detection. S pilin (14 kDa)
cannot be visualized on panel B, since the characterization of
significant differences in the migration of glycosylated and
nonglycosylated full-length monomers of pilin requires long
electrophoresis periods, such that truncated S pilin migrates off the
gel. On the other hand, S pilin can be seen after a short migration
period (panel A, lanes 1G, 2M, and 2G); however, no molecular weight
difference in the full-length monomer can be appreciated. Lanes 1E and
2E show two different products of the full-length pilin gene: a large
one, which is prepilin, whose signal peptide is not cleaved, and a
smaller one, which corresponds to mature pilin. As previously shown
(3), this result is due to the lack of prepilin peptidase,
resulting in incomplete processing of prepilin.
|
|
The production of S pilin by bacteria expressing modified or
nonmodified pilin variants was compared. As already mentioned,
S pilin
was not produced by an MC derivative expressing the parental
nonglycosylated RM11 variant (Fig.
5,
lane 3M) (
RS, 0), whereas
a GC strain expressing
the same pilin produced a significant amount
of S pilin (Fig.
5, lane
3G) (
RS, 0.54 ± 0.11). On the other hand,
the RM11rg derivative produced S pilin even when expressed in
MC (Fig.
5, lane 4M) (
RS, 0.48 ± 0.06) but in an
amount smaller
than that produced in GC (Fig.
5, lane 4G)
(
RS, 0.7 ± 0.09). These
data show that the
posttranslational modification introduced in
RM11rg favors the
production of S pilin in both GC and MC.
View larger version (25K):
[in this window]
[in a new window]
|
FIG. 5.
Western blot of total bacterial extracts. Samples were
electrophoresed by SDS-15% PAGE. Lanes 1, SA-producing strains
(glycosylated pilin); lanes 2, SA(Ser62Ala-Ser63Ala)-producing strains
(nonglycosylated pilin); lanes 3, RM11-producing strains
(nonglycosylated pilin); lanes 4, RM11rg-producing strains
(glycosylated pilin). Total protein extracts of MC strain 8013 or GC
strain FA1090 expressing the different pilin variants are designated by
M or G, respectively. The values below the lanes correspond to the mean
(standard deviation) RS values determined by
density analysis in at least four independent experiments containing
various quantities of proteins.
|
|
In order to confirm that a nonglycosylated pilin variant was
capable of producing S pilin in GC, the SA(Ser62Ala-Ser63Ala)
variant of SA, which has two Ala residues in place of Ser62 and
Ser63 and which has been shown to be nonglycosylated, was introduced
into GC strain FA1090 (
14). The amount of S pilin produced
by
the parental glycosylated SA pilin was more abundant in GC than
in
MC (Fig.
5, lanes 1M and 1G) (
RS, 0.23 ± 0.06 and 0.57 ± 0.04,
respectively). When similar strains
expressed the nonglycosylated
variant, no S pilin was produced in MC
(Fig.
5, lane 2M) (
RS,
0), whereas pilin was
still processed as a truncated soluble form
in GC (Fig.
5, lane 2G)
(
RS, 0.28 ± 0.08). These data confirmed
that in GC, as in MC, pilin glycosylation favors the production
of S
pilin. However, the fact that the loss of glycosylation did
not
entirely prevent the processing of pilin into S pilin demonstrates
that
in GC, unlike in MC, glycosylation is not required for S-pilin
production and suggests that the mechanisms responsible for the
production of S pilin in MC and GC are
different.
| |
DISCUSSION |
Pilus-mediated adhesion is one of the major mechanism by which GC
and MC can interact with cells. For encapsulated MC, it is the only
means by which bacteria can adhere to cells. Pilin plays a major role
in this process by being the scaffold for pili. The cell-binding domain
is thought to be carried by the PilC proteins. However, in both MC and
GC, pilin antigenic variation has been shown to be responsible for
intrastrain variability in bacterial adhesiveness. The mechanism of
this variability has been explored with MC, and high-adhesion pilin has
been shown to be responsible for the formation of bundles of pili,
which promote the interbacterial interactions responsible for the
growth of microorganisms as colonies on infected monolayers. In order
to address whether a similar mechanism is responsible for intrastrain
variability in GC, we introduced low- and high-adhesion MC pilin
variants into GC and low- and high-adhesion variants isolated from GC
into MC. To achieve this goal, transcriptional fusions were created
between pilin and a Kmr resistance gene and then were
shuttled back into MC and GC strains. Thus, all variants studied were
introduced into the same backgrounds.
Our data demonstrated that (i) in GC, as in MC, high-adhesion pilin
variants form bundles of pili and (ii) each variant studied is
responsible for equivalent pilus-mediated adhesiveness and identical
pilus morphology in both species, thus showing that full-size pilins
are functionally exchangeable between GC and MC. It should be pointed
out that strains expressing pilins from pilE::Km
fusions were always less adhesive than wild-type strains producing the
same variants. The most likely explanation for this finding is a
modification of the half-life of the pilin mRNA induced by the
transcriptional fusion, thus leading to a decreased concentration of
pilin. In addition, this hypothesis is consistent with the fact that
RM11 in the wild type was clearly piliated, whereas a strain expressing
RM11 fused to the Kmr gene was not piliated.
During the course of these experiments, we observed that
S-pilin-producing variants make more truncated monomers in GC than in
MC. In a previous work, glycosylation was shown to be required for the
production of S pilin in MC (14), and the glycosylation site
was identified as Ser63. Replacement of this residue by an Ala in SA
abolished the ability of an MC strain to produce S pilin. On the other
hand, a GC strain expressing a nonglycosylated derivative of SA was
still capable of producing a significant amount of S pilin. Similar
results were obtained with the RM11 pilin variant. This latter pilin
does not have Ser63; furthermore, its electrophoretic mobility is
identical in GC or E. coli, thus eliminating the possibility that it is glycosylated at another site. Consistent with the above data, S pilin is produced by this variant only in GC and not in MC. On
the other hand, restoration of glycosylation by introduction of Ser63
induces the production of S pilin in MC. The mechanism by which
glycosylation favors S-pilin production is unknown. However, protein
glycosylation is well known to help in solubilizing proteins, and a
nonspecific effect of this posttranslational modification is likely.
The reason why nonglycosylated GC pilin remained capable of producing S
pilin is unknown. An additional glycosylation site in GC pilin was not
a likely explanation, as there was no detectable difference in
molecular weight between the products of both species with identical
pilE genes. On the other hand, pilin has been shown to
undergo several posttranslational modifications (14, 25); one of these could be specific for pilin produced in GC and could be
responsible for the difference in pilin solubilization.
No function has been found yet for S pilin. It was recently shown that
there is a cell-binding domain within the first 77 residues of mature
pilin (13). The structural data suggest that this potential
pilin cell-binding domain is hidden in assembled pili. One hypothesis
is that it may be available for binding only in the context of the
soluble monomer. This cell-binding domain could subsequently play a
role in meningococcal pathogenesis by interacting with host membrane or
immune system components. Work is currently in progress to test this hypothesis.
| |
ACKNOWLEDGMENTS |
We thank Hank Seifert for providing us with strains and C. Tinsley for careful reading of the manuscript.
M.M. was the recipient of a scholarship from the Ministère de la
Recherche et de la Technologie. X.N. was supported by INSERM, Université René Descartes, the DRET, and the Fondation pour la Recherche Médicale.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: INSERM U411,
Faculté de Médecine Necker-Enfants Malades, 156 rue de
Vaugirard, 75015 Paris, France. Phone: 33-140615375. Fax: 33-140615592. E-mail: nassif{at}necker.fr.
| |
REFERENCES |
| 1.
|
Achtman, M.,
M. Neibert,
B. A. Crowe,
W. Strittmatter,
B. Kusecek,
E. Weyse,
M. J. Walsh,
B. Slawig,
G. Morelli,
A. Moll, and M. Blake.
1988.
Purification and characterization of eight class 5 outer membrane protein variants from a clone of Neisseria meningitidis serogroup A.
J. Exp. Med.
168:507-525[Abstract/Free Full Text].
|
| 2.
|
Connel, T. D.,
W. J. Black,
K. T. H.,
D. S. Barritt,
J. A. Dempsey,
K. Kverneland, Jr.,
A. Stephenson,
G. L. Schepart,
G. L. Murphy, and J. G. Cannon.
1988.
Recombination among protein II genes of Neisseria gonorrhoeae generates new coding sequences and increases structural variability in the protein II family.
Mol. Microbiol.
2:227-236[Medline].
|
| 3.
|
Dupuy, B.,
M. K. Taha,
A. P. Pugsley, and C. Marchal.
1991.
Neisseria gonorrhoeae prepilin export studied in Escherichia coli.
J. Bacteriol.
173:7589-7598[Abstract/Free Full Text].
|
| 4.
|
Guibourdenche, M.,
M. Y. Popoff, and J. Y. Riou.
1986.
Deoxyribonucleic acid relatedness among Neisseria gonorrhoeae, N. meningitidis, N. lactamica, N. cinerea and Neisseria polysaccharea.
Ann. Inst. Pasteur/Microbiol. (Paris)
137B:177-185.
|
| 5.
|
Haas, R.,
H. Schwarz, and T. Meyer.
1987.
Release of soluble pilin antigen coupled with gene conversion in Neisseria gonorrhoeae.
Proc. Natl. Acad. Sci. USA
84:9079-9083[Abstract/Free Full Text].
|
| 6.
|
Hagblom, P.,
E. Segal,
E. Billyard, and M. So.
1985.
Intragenic recombination leads to Neisseria gonorrhoeae pilus antigenic variation.
Nature
315:156-158[Medline].
|
| 7.
|
Ho, S. N.,
H. D. Hunt,
R. M. Horton,
J. K. Pullen, and L. R. Pease.
1989.
Site-directed mutagenesis by overlap extension using the polymerase chain reaction.
Gene
77:51-59[Medline].
|
| 8.
|
Jonsson, A.-B.,
G. Nyberg, and S. Normark.
1991.
Phase variation of gonococcal pili by frameshift mutation in pilC, a novel gene for pilus assembly.
EMBO J.
10:477-488[Medline].
|
| 9.
|
Kellog, D. D.,
W. L. Peacock,
W. E. Deacon,
L. Brown, and C. I. Pirkle.
1963.
Neisseria gonorrhoeae. I. Virulence genetically linked to a clonal variation.
J. Bacteriol.
141:1274-1279.
|
| 10.
|
Laemmli, U. K.
1970.
Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature
227:680-685[Medline].
|
| 11.
|
Lambden, P. R.,
J. N. Robertson, and P. J. Watt.
1980.
Biological properties of two distinct pilus types produced by isogenic variants of Neisseria gonorrhoeae P9.
J. Bacteriol.
141:393-396[Abstract/Free Full Text].
|
| 12.
|
Long, C. D.,
R. N. Madraswala, and H. S. Seifert.
1998.
Comparisons between colony phase variation of Neisseria gonorrhoeae FA1090 and pilus, pilin, and S-pilin expression.
Infect. Immun.
66:1918-1927[Abstract/Free Full Text].
|
| 13.
|
Marceau, M.,
J.-L. Beretti, and X. Nassif.
1995.
High adhesiveness of encapsulated Neisseria meningitidis to epithelial cells is associated with the formation of bundles of pili.
Mol. Microbiol.
17:855-863[Medline].
|
| 14.
|
Marceau, M.,
K. Forest,
J. L. Béretti,
J. Tainer, and X. Nassif.
1997.
Consequences of the loss of O-linked glycosylation of meningococcal type IV pilin on piliation and pilus mediated adhesion.
Mol. Microbiol.
27:705-716.
|
| 15.
|
Nassif, X.,
J.-L. Beretti,
J. Lowy,
P. Stenberg,
P. O'Gaora,
J. Pfeifer,
S. Normark, and M. So.
1994.
Roles of pilin and PilC in adhesion of Neisseria meningitidis to human epithelial and endothelial cells.
Proc. Natl. Acad. Sci. USA
91:3769-3773[Abstract/Free Full Text].
|
| 16.
|
Nassif, X.,
J. Lowy,
P. Stenberg,
P. O'Gaora,
A. Ganji, and M. So.
1993.
Antigenic variation of pilin regulates adhesion of Neisseria meningitidis to human epithelial cells.
Mol. Microbiol.
8:719-725[Medline].
|
| 17.
|
Nassif, X.,
D. Puaoi, and M. So.
1991.
Transposition of Tn1545- 3 in the pathogenic neisseriae: a genetic tool for mutagenesis.
J. Bacteriol.
173:2147-2154[Abstract/Free Full Text].
|
| 18.
|
Olafson, R. W.,
P. J. McCarthy,
A. R. Bhatti,
J. S. G. Dooley,
J. E. Heckels, and T. J. Trust.
1985.
Structural and antigenic analysis of meningococcal piliation.
Infect. Immun.
48:336-342[Abstract/Free Full Text].
|
| 19.
|
Parge, H. E.,
K. T. Forest,
M. J. Hickey,
D. Christensen,
E. D. Getzoff, and J. A. Tainer.
1995.
Structure of the fibre-forming protein pilin at 2.6 A resolution.
Nature
378:32-38[Medline].
|
| 20.
|
Potts, W. J., and J. R. Saunders.
1988.
Nucleotide sequence of the structural gene for class I pilin from Neisseria meningitidis: homologies with the pilE locus of Neisseria gonorrhoeae.
Mol. Microbiol.
2:647-653[Medline].
|
| 21.
|
Rahman, M.,
M. Kallstrom,
S. Normark, and A. B. Jonsson.
1997.
PilC of pathogenic Neisseria is associated with the bacterial surface.
Mol. Microbiol.
25:11-25[Medline].
|
| 22.
|
Rudel, T.,
I. Scheuerpflug, and T. F. Meyer.
1995.
Neisseria PilC protein identified as type-4 pilus tip-located adhesin.
Nature
373:357-362[Medline].
|
| 23.
|
Rudel, T.,
J. P. M. Van Putten,
C. P. Gibbs,
R. Haas, and T. F. Meyer.
1992.
Interaction of two variable proteins (PilE and PilC) required for pilus-mediated adherence of Neisseria gonorrhoeae to human epithelial cells.
Mol. Microbiol.
6:3439-3450[Medline].
|
| 24.
|
Sambrook, J.,
E. F. Fritsch, and T. Maniatis.
1989.
Molecular cloning: a laboratory manual, 2nd ed.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
|
| 25.
|
Stimson, E.,
M. Virji,
S. Barker,
M. Panico,
I. Blench,
J. R. Saunders,
G. Payne,
E. R. Moxon,
A. Dell, and H. R. Morris.
1996.
Discovery of a novel protein modification: alpha-glycerophosphate is a substituent of meningococcal pilin.
Biochem. J.
316:29-33.
|
| 26.
|
Stimson, E.,
M. Virji,
K. Makepeace,
A. Dell,
H. R. Morris,
G. Payne,
J. R. Saunders,
M. P. Jennings,
S. Barker,
M. Panico,
I. Blench, and E. R. Moxon.
1995.
Meningococcal pilin: a glycoprotein substituted with digalactoside 2,4-diacetamido-2,4,6-trideoxyhexose.
Mol. Microbiol.
17:1201-1214[Medline].
|
| 27.
|
Swanson, J.,
E. Sparks,
B. Zeligs,
M. A. Siam, and C. Parrott.
1974.
Studies on gonococcus infection. V. Observations on in vitro interactions of gonococci and human neutrophils.
Infect. Immun.
10:633-644[Abstract/Free Full Text].
|
| 28.
|
Virji, M.,
C. Alexandrescu,
D. J. P. Fergusson,
J. R. Saunders, and E. R. Moxon.
1992.
Variations in the expression of pili: the effect on adherence of Neisseria meningitidis to human epithelial and endothelial cells.
Mol. Microbiol.
6:1271-1279[Medline].
|
| 29.
|
Virji, M.,
J. E. Heckels,
W. H. Potts,
C. A. Hart, and J. R. Saunders.
1989.
Identification of epitopes recognised by monoclonal antibodies SM1 and SM2 which react with all pili of Neisseria gonorrhoeae but which differentiate between two structural classes of pili expressed by Neisseria meningitidis and the distribution of their encoding sequences in the genomes of Neisseria sp.
J. Gen. Microbiol.
135:3239-3251[Abstract/Free Full Text].
|
| 30.
|
Virji, M.,
J. R. Saunders,
G. Sims,
K. Makepeace,
D. Maskell, and D. J. P. Ferguson.
1993.
Pilus-facilitated adherence of Neisseria meningitidis to human epithelial and endothelial cells: modulation of adherence phenotype occurs concurrently with changes in primary amino acid sequence and the glycosylation status of pilin.
Mol. Microbiol.
10:1013-1028[Medline].
|
Journal of Bacteriology, January 1999, p. 656-661, Vol. 181, No. 2
0021-9193/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Kus, J. V., Kelly, J., Tessier, L., Harvey, H., Cvitkovitch, D. G., Burrows, L. L.
(2008). Modification of Pseudomonas aeruginosa Pa5196 Type IV Pilins at Multiple Sites with D-Araf by a Novel GT-C Family Arabinosyltransferase, TfpW. J. Bacteriol.
190: 7464-7478
[Abstract]
[Full Text]
-
Asikyan, M. L., Kus, J. V., Burrows, L. L.
(2008). Novel Proteins That Modulate Type IV Pilus Retraction Dynamics in Pseudomonas aeruginosa. J. Bacteriol.
190: 7022-7034
[Abstract]
[Full Text]
-
Taguchi, F., Shibata, S., Suzuki, T., Ogawa, Y., Aizawa, S.-I., Takeuchi, K., Ichinose, Y.
(2008). Effects of Glycosylation on Swimming Ability and Flagellar Polymorphic Transformation in Pseudomonas syringae pv. tabaci 6605. J. Bacteriol.
190: 764-768
[Abstract]
[Full Text]
-
Voisin, S., Kus, J. V., Houliston, S., St-Michael, F., Watson, D., Cvitkovitch, D. G., Kelly, J., Brisson, J.-R., Burrows, L. L.
(2007). Glycosylation of Pseudomonas aeruginosa Strain Pa5196 Type IV Pilins with Mycobacterium-Like {alpha}-1,5-Linked D-Araf Oligosaccharides. J. Bacteriol.
189: 151-159
[Abstract]
[Full Text]
-
Karlyshev, A. V., Everest, P., Linton, D., Cawthraw, S., Newell, D. G., Wren, B. W.
(2004). The Campylobacter jejuni general glycosylation system is important for attachment to human epithelial cells and in the colonization of chicks. Microbiology
150: 1957-1964
[Abstract]
[Full Text]
-
Kus, J. V., Tullis, E., Cvitkovitch, D. G., Burrows, L. L.
(2004). Significant differences in type IV pilin allele distribution among Pseudomonas aeruginosa isolates from cystic fibrosis (CF) versus non-CF patients. Microbiology
150: 1315-1326
[Abstract]
[Full Text]
-
Young, N. M., Brisson, J.-R., Kelly, J., Watson, D. C., Tessier, L., Lanthier, P. H., Jarrell, H. C., Cadotte, N., St. Michael, F., Aberg, E., Szymanski, C. M.
(2002). Structure of the N-Linked Glycan Present on Multiple Glycoproteins in the Gram-negative Bacterium, Campylobacter jejuni. J. Biol. Chem.
277: 42530-42539
[Abstract]
[Full Text]
-
Banerjee, A., Wang, R., Supernavage, S. L., Ghosh, S. K., Parker, J., Ganesh, N. F., Wang, P. G., Gulati, S., Rice, P. A.
(2002). Implications of Phase Variation of a Gene (pgtA) Encoding a Pilin Galactosyl Transferase in Gonococcal Pathogenesis. JEM
196: 147-162
[Abstract]
[Full Text]
-
Szymanski, C. M., Burr, D. H., Guerry, P.
(2002). Campylobacter Protein Glycosylation Affects Host Cell Interactions. Infect. Immun.
70: 2242-2244
[Abstract]
[Full Text]
-
Rytkonen, A., Johansson, L., Asp, V., Albiger, B., Jonsson, A.-B.
(2001). Soluble Pilin of Neisseria gonorrhoeae Interacts with Human Target Cells and Tissue. Infect. Immun.
69: 6419-6426
[Abstract]
[Full Text]
-
Kahler, C. M., Martin, L. E., Tzeng, Y.-L., Miller, Y. K., Sharkey, K., Stephens, D. S., Davies, J. K.
(2001). Polymorphisms in Pilin Glycosylation Locus of Neisseria meningitidis Expressing Class II Pili. Infect. Immun.
69: 3597-3604
[Abstract]
[Full Text]
-
Hamrick, T. S., Dempsey, J. A. F., Cohen, M. S., Cannon, J. G.
(2001). Antigenic variation of gonococcal pilin expression in vivo: analysis of the strain FA1090 pilin repertoire and identification of the pilS gene copies recombining with pilE during experimental human infection. Microbiology
147: 839-849
[Abstract]
[Full Text]
-
Hui Wu, , Fives-Taylor, P. M.
(2001). Molecular Strategies for Fimbrial Expression and Assembly. CROBM
12: 101-115
[Abstract]
[Full Text]
-
Power, P. M., Roddam, L. F., Dieckelmann, M., Srikhanta, Y. N., Cheng Tan, Y., Berrington, A. W., Jennings, M. P.
(2000). Genetic characterization of pilin glycosylation in Neisseria meningitidis. Microbiology
146: 967-979
[Abstract]
[Full Text]
Top
Abstract
Introduction
