Periodontology 2000, Vol.

35, 2004, 101±134 Copyright # Blackwell Munksgaard 2004

Printed in Denmark. All rights reserved PERIODONTOLOGY 2000

Antigens of bacteria associated

with periodontitis
Neil M. O'Brien-Simpson, Paul D. Veith, Stuart G. Dashper &
Eric C. Reynolds

From detailed epidemiological research and animal

models of disease, Actinobacillus actinomycetemco- Identification of periodontal
mitans, Porphyromonas gingivalis, Tannerella forsy- pathogens
thia (formerly Bacteroides forsythus) and Treponema
denticola have been implicated, either singly or in During the 1960s and 70s a number of studies indi-
combination, in the development of various forms of cated that chronic periodontitis was associated with
periodontitis. These four species are all Gram-nega- speci®c bacteria in subgingival plaque (reviewed in
tive anaerobic bacteria. Using genomic, proteomic Page 1999 (264)). This led to the development of the
and immunological strategies, antigens of these speci®c plaque hypothesis where the on-set and pro-
pathogens are rapidly being identi®ed. An extensive gression of periodontitis was suggested to be asso-
review of the literature revealed that lipopolysac- ciated with speci®c bacterial pathogens (196, 197).
charide or outer membrane lipids, polysaccharide, Support for this hypothesis has grown substantially
®mbriae and outer membrane and secreted pro- since these early studies, for example, Slots (325) and
teins are antigens of all four bacteria that may play Newman et al. (241) suggested that A. actinomyce-
a role in disease pathogenesis. However, a feature temcomitans was the major bacterium associated
that may be important in differentiating these with localized juvenile/early-onset periodontitis
microorganisms from many of the other Gram- (aggressive periodontitis (5)) and Tanner et al. (350)
negative subgingival plaque bacterial species is suggested the four bacterial species; P. gingivalis, T.
extracellular proteolytic activity. These enzymes forsythia (formerly B. forsythus (298)), A. actinomy-
have the potential to degrade host matrix proteins cetemcomitans and Eikenella corrodens were the
and host matrix proteinase inhibitors while activat- major pathogens associated with adult periodontitis
ing latent host matrix metalloproteinases, induce (chronic periodontitis (5)). From a large number of
pro-in¯ammatory cytokines involved in tissue des- subsequent studies (reviewed in Haffajee and
truction and alveolar bone resorption and dysregu- Socransky (103) and Zambon (402)) a consensus
late host defence by degrading inter alia cellular report of the World Workshop on Clinical Periodontics
receptors, complement, antibodies and certain cyto- in 1996 (39) concluded that three bacterial species
kines. Studies using animal models of disease with (P. gingivalis, T. forsythia, A. actinomycetemcomitans)
isogenic mutants of two of these pathogens lacking were not only associated with, but were the aetiologi-
speci®c proteolytic activity are consistent with the cal agents of, periodontitis. As well as these three
extracellular proteinases being major virulence bacteria there was moderate evidence to associate a
factors. The signi®cant reduction of periodontal dis- number of other bacteria with periodontitis (T. denti-
ease progression in animal models by immuniza- cola, Campylobacter rectus, Eubacterium nodatum,
tion with bacterial virulence factors, in particular Fusobacterium nucleatum, Prevotella intermedia/
the proteinases, suggests that a multispecies vac- nigrescens, Peptostreptococcus micros and Streptococ-
cine may be an important adjunctive therapy to cus intermedius). The report recommended that
mechanical debridement in the treatment of peri- molecular probe technology be used to improve
odontitis in humans. our understanding of the role of these and other

101
O'Brien-Simpson et al.

non-cultivable bacteria in disease aetiology. Using more bone loss than that found in uninoculated ani-
checkerboard DNA±DNA hybridization analysis, mals. Furthermore, subgingival implantation of
Socransky et al. (328) determined the levels of 40 sub- A. actinomycetemcomitans in mice and rats has been
gingival taxa in 13,261 plaque samples from chronic shown to induce 30±60% more horizontal bone loss
periodontitis patients. The authors were able to group than that of uninoculated animals (352, 391).
these taxa into bacterial communities (complexes) Although over twenty bacterial species have been
using clustering and ordination techniques. The bac- associated with periodontitis, only four species
terial taxa could be grouped into 5 major complexes P. gingivalis, T. forsythia, T. denticola and A. actino-
with one complex in particular, termed the red com- mycetemcomitans have been consistently and
plex, comprising P. gingivalis, T. denticola and T. for- strongly associated with progression of disease
sythia, being highly associated with the clinical (327, 328, 402). This, together with their virulence
measures of chronic periodontitis, especially pocket in animal models, suggests that these four bacterial
depth and bleeding on probing. This study con®rmed species may be the major pathogens associated with
earlier ®ndings that members of the red complex were periodontitis in humans. This review will focus on
regularly found in subgingival plaque from chronic the identi®cation and role in virulence of antigens
periodontitis patients. For example, a strong relation- from these four bacterial pathogens.
ship between disease severity and the presence of
P. gingivalis with T. forsythia (87) and P. gingivalis
with T. denticola (317) in subgingival plaque had pre- Methods of identification of
viously been reported. In both studies these bacteria bacterial protein antigens and
were found more frequently, and in higher numbers, epitopes
in deeper periodontal pockets with probing depths of
>4 mm.
Identification of antigens by proteomic
A number of studies have shown that P. gingivalis
analysis
co-aggregates with T. denticola (94, 262) and other
oral bacterial species such as Fusobacterium spp. and As antigens are very often surface-exposed proteins,
Actinomyces spp. (62, 63, 157, 167) but not with one strategy for their identi®cation is to characterize
A. actinomycetemcomitans (168). T. denticola co- the proteins present in surface extracts of the patho-
aggregates with Fusobacterium spp. but not with gen of interest. Often, such a strategy involves ana-
A. actinomycetemcomitans and Actinomyces spp. lysis of the surface extract by Western blots of
(169). Bacterial co-aggregation is important for the one-dimensional (1D) Sodium Dodecyl Sulphate
development of bio®lms such as subgingival plaque Polyacrylamide Gel Electrophoresis (SDS±PAGE) gels
and colonization of tissue surfaces (158). The virulence to identify antigenic proteins, which subsequently
of periodontal pathogens has been studied using a may be identi®ed by N-terminal sequencing by
number of animal lesion/abscess models. Mono- Edman degradation. Identi®cation of SDS±PAGE
infection, by subcutaneous injection, of T. forsythia, separated proteins by N-terminal sequencing alone,
T. denticola, A. actinomycetemcomitans and P. gingi- however can be problematic. Surface-associated pro-
valis non-invasive strains, ATCC 33277 and 381, (240, teins are often heavily glycosylated (220), and several
363) produce localized abscesses at the site of other kinds of modi®cation, such as acetylation, for-
injection (18, 30, 57, 151). However, P. gingivalis mylation and pyroglutamine formation, are known
invasive strains, W50 and A7A1-28 (ATCC 53977), to affect the N-terminus of proteins in such a way
produce spreading ulcerative lesions at, and distant that usually precludes straightforward analysis by
from, the site of injection (240, 363). In the N-terminal sequencing (114). Proteins derived
murine lesion model co-infection of P. gingivalis±T. from complex samples such as preparations of bac-
denticola, P. gingivalis±T. forsythia, P. gingivalis± terial cell walls and membranes cannot be ade-
F. nucleatum and P. gingivalis±A. actinomycetemco- quately resolved by 1D SDS±PAGE alone. Thus
mitans produced larger lesions and enhanced although sometimes successful in identifying anti-
virulence compared with mono-infection of each of gens from puri®ed or semi-purifed preparations,
the bacterial species (32, 59, 152, 397). In animal these techniques have now been surpassed by pro-
models of periodontitis, subgingival implantation of teomic analysis which can rapidly identify protein
P. gingivalis in mice (9), rats (65, 144) and non-human antigens from complex mixtures, especially for
primates (118, 271) induces the progression of those organisms whose genomes have been seq-
periodontal bone loss, causing between 50±110% uenced. The genomes of P. gingivalis, T. denticola,

102
 Antigens of bacteria associated with periodontitis

and A. actinomycetemcomitans can be downloaded and C-termini and modi®cation sites can often be
from the following website addresses: [Link] obtained.
[Link]/tigr-scripts/CMR2/[Link], http:// Proteomic techniques have been used to charac-
[Link]/microbial/Tdenticola/ and http:// terize surface or secreted proteins from a number of
[Link]/[Link], respectively, and bacteria including Escherichia coli (226), Helicobacter
searched using the Basic Local Alignment Search Tool pylori (23, 242), Borrelia garanii (138), Caulobacter
(BLAST) engine. The genome of T. forsythia (listed as crescentus (273), Pseudomonas aeruginosa (247) and
B. forsythus) is currently being sequenced by The Insti- Mycobacterium tuberculosis (378). In C. crescentus for
tute for Genomic Research (TIGR) and can be searched example, 41 outer membrane proteins (OMPs) were
using the BLAST search engine at the TIGR website identi®ed in a single study employing 2D-PAGE and
[Link] PMF (273), and in B. garanii, 65 antigens were detec-
The proteomic approach of using two dimensional ted on 2D-Western blots from which 13 were identi-
(2D)-PAGE separation and mass spectrometric tech- ®ed by speci®c antisera, 4 were identi®ed by PMF
niques to identify the proteins overcomes the pro- and a further 3 identi®ed by both speci®c antisera
blems associated with 1D SDS±PAGE separation of and PMF (138). Except for the work on P. gingivalis
complex samples (reviewed by Humphery-Smith outlined below, very little use of proteomic techniques
et al. (122)). 2D-PAGE is capable of resolving several has been made in the study of periodontal pathogens.
thousand proteins on a single gel by enabling them However, the completion of the sequencing of the
to be separated over two dimensions (281). The ®rst A. actinomycetemcomitans, T. forsythia and T. denti-
dimension used is most frequently isoelectric focuss- cola genomes is likely to result in a great increase in
ing (IEF) which separates proteins by charge, while antigen identi®cation over the next few years.
the second dimension is SDS±PAGE that separates
proteins according to their molecular size. After the
Identification of B-cell and T-cell
second dimension, the 2D-gel can be subjected to
antigenic determinants using epitope
Western blotting, while a second 2D-gel can be
mapping analysis
stained, and the protein spots subjected to mass
spectrometric (MS) analysis. For MS analysis, a pro- Once an antigenic protein has been identi®ed and
tease (often trypsin) is added to the gel-spot to cleave the amino acid sequence deduced using proteomic
the protein into peptide fragments, these peptides and genomic techniques, those parts (antigenic epi-
are subsequently recovered and analysed by an MS topes) that are important in the host immune
technique known as peptide mass ®ngerprinting response to the protein can be identi®ed using a
(PMF) (112). PMF generates a list of peptide masses number of different techniques. Computer algo-
that can be used to identify the protein by reference rithms are one technique that can be used to predict
to a peptide mass database derived from all possible potential B-cell and T-cell antigenic epitopes by ana-
protein sequences encoded by the pathogen's gen- lysing the amino acid sequence of a given protein.
ome. The advantages of using PMF to identify pro- B-cell epitopes can be predicted by analysing hydro-
teins include its speed and sensitivity. In addition to philic, hydrophobic, segment mobility, a-helical
the detection of more proteins, PMF has the further and b-turn segments within a protein (11, 121, 178,
advantage of detecting proteins that co-migrate in a 248, 379) and assigning antigenic index values for
single spot in the gel. When N-terminal sequencing amino acid sequences (134). Algorithms predicting
by Edman degradation is employed, it is often very T-cell epitopes use structural data from MHC±pep-
dif®cult to obtain accurate sequence when two or tide complexes and peptide±MHC binding data to
more proteins are present in the one sample, espe- analyse the amino acid sequence of a given protein
cially if the proteins are present in similar amounts. and predict those peptide sequences that may bind to
Moreover, if the mixture contains N-terminally a particular MHC molecule. A number of computer
blocked proteins as well as a single unblocked pro- algorithms are available on the world wide web to
tein, Edman sequencing will give the false impres- predict cytotoxic T-cell (CD8‡) MHC-class I epitopes,
sion that the sample contains a pure protein, even e.g `BIMAS' developed by Parker et al. (267) (http://
if the unblocked protein is the least abundant. PMF [Link]/molbio/hla_bind/). To predict
on the other hand is capable of identifying several T-helper cell (CD4‡) MHC±class II epitopes the
co-migrating proteins in the one spot. As PMF software package `TEPITOPE' developed by Sturniolo
data veri®es the sequence from any part of the pro- et al. (331) can be downloaded from the VACCINOME
tein, information regarding the location of the N- website [Link]

103
O'Brien-Simpson et al.

Another computer prediction program available over anaerobic coccobacillus. A. actinomycetemcomitans

the Internet is `SYFPEITHI' developed by Rammensee can be classi®ed into six distinct serotypes (a±f)
et al. (283) which can be used to predict MHC-class I based on surface polysaccharides located on the O
and class II T-cell epitopes ([Link] side chains of lipopolysaccharide (88, 140, 295, 404).
[Link]/syfpeithi/). Although computer algorithms The chemical structures of these polysaccharides
have been successfully used to identify B-cell and have been determined and have been shown to con-
T-cell epitopes (134, 177, 202, 225) the predicted seq- sist of repeating disaccharide units of O-acetyl-6-
uences still need to be tested in bio-assays to evalu- deoxy-d-talose and 6-deoxy-d-talose (serotype a);
ate whether they bind antibody or stimulate T-cells. O-acetyl-6-deoxy-l-talose and 6-deoxy-l-talose (ser-
Limitations of computer algorithms in predicting otype c); 2-acetamido-2-deoxy-d-glucose and l-
antigenic epitopes centre around a limited data base rhamnose (serotype e), repeating trisaccharide units
to make the predictions from, although the rapidly of l-rhamnose, d-fucose and N-acetyl-d-galactosa-
expanding information on protein structure and mine (serotype b); 2-acetamido-2-deoxy-d-galactose
function will enhance epitope prediction algorithms. and l-rhamnose (serotype f) or repeating tetrasac-
A systematic method for identifying both B-cell epi- charide units of d-glucose, d-mannose and l-rham-
topes and T-cell epitopes is PEPSCAN (84, 207, 362). nose (serotype d) (3, 140, 269, 270). The presence of
The principle of the method is to synthesize over- A. actinomycetemcomitans in subgingival plaque has
lapping peptides of a given length (e.g. 8 to 15 mers) been associated with aggressive periodontitis with
covering the whole or part of a given protein seq- serotype b twice as prevalent as serotype a (402,
uence onto a solid support. For B-cell epitope identi- 404). These data are supported by a number of other
®cation the amino acid side chain protecting groups studies that have shown serotype b to be the predo-
used in peptide synthesis are removed and the pep- minant A. actinomycetemcomitans serotype asso-
tides are left on the solid support. The solid support- ciated with periodontitis (25, 54, 83, 86, 100, 195).
bound peptides are then probed in an ELISA with Animal studies have shown that serotype b induces
antisera to the whole organism or protein under the pro-in¯ammatory cytokine interleukin-6 (IL-6)
investigation. For identi®cation of T-cell epitopes from murine splenocytes (143). Furthermore, sero-
the peptides need to be cleaved from the solid sup- type b induces more IL-1 (another pro-in¯ammatory
port and used in cytotoxic T-cell or T-cell prolifera- cytokine) from thymocytes than serotype a or c (343).
tion assays with bacteria- or protein-primed T-cells. Ebersole et al. (57) have reported that the order of
Although the PEPSCAN approach has been applied to virulence for A. actinomycetemcomitans serotypes in
identify epitopes on an A. actinomycetemcomitans the murine lesion model was serotype b > a > c. The
antigen (183), it has been mostly applied in identifying serotype-speci®c polysaccharides are reported to be
P. gingivalis antigenic epitopes and is discussed below. major antigenic targets of the immune response of
The identi®cation of antigenic epitopes of a parti- aggressive periodontitis patients infected with
cular protein from a pathogenic bacterium forms an A. actinomycetemcomitans (25, 319) with the predo-
essential part of the development and design of sub- minant antibody subclass being IgG2 followed by
unit/synthetic vaccines and immunodiagnostics. IgG1 and IgG3 (193, 194, 203, 384). Gu et al. (99) have
This knowledge also enhances our understanding shown that sera from localized aggressive periodon-
of the role a particular protein antigen plays in the titis patients as well as binding to O-polysaccharide
host immune response. Thus by combining proteo- recognize other components of lipopolysaccharide
mic, genomic and epitope mapping techniques such as lipid A and the core polysaccharide a-keto-
together with animal and human studies it is possible 3-deoxyoctonate moieties.
to identify pathogen antigens and elucidate their A. actinomycetemcomitans serotype b lipopoly-
immunological and biological role in disease. saccharide and lipid A induce the release of the
pro-in¯ammatory cytokine IL-1 from murine peri-
toneal macrophages and cell line macrophages
Actinobacillus (244). Furthermore A. actinomycetemcomitans lipo-
actinomycetemcomitans antigens polysaccharide also stimulated the release of IL-1b
from polymorphonuclear leukocytes as well as
other pro-in¯ammatory cytokines, such as, tumor
Polysaccharide and lipopolysaccharide
necrosis factor-a (TNF-a) and IL-8 (400). Murine
A. actinomycetemcomitans is a gram-negative, non- bone marrow cells stimulated with lipopolysac-
spore forming, non-motile, capnophilic, facultative charide from A. actinomycetemcomitans have also

104
 Antigens of bacteria associated with periodontitis

been reported to release IL-1a and prostaglandin E2 cells, is suggested to facilitate insertion into the cell
(PGE2) (355). membrane and pore formation (184). Korostoff et al.
Pro-in¯ammatory cytokines such as IL-1a, IL-1b (170) have shown that A. actinomycetemcomitans
and TNF-a induce bone resorption by promoting leukotoxin at low doses induces apoptosis in HL-60
differentiation of osteoclast precursor cells and by cells, which supported earlier observations that cell
activating osteoclast cells (272). Ito et al. (131) have death of T-cells and NK cells caused by leukotoxin
reported that lipopolysaccharide from A. actinomy- was by an apoptotic mechanism (214, 312). As at high
cetemcomitans promotes differentiation of osteoclasts doses, leukotoxin has been reported to cause cell
in vitro in the presence of 1,25-dihydroxyvitamin D3 death by necrosis, then it has been suggested that
and dexamethasone. Other studies have shown that A. actinomycetemcomitans leukotoxin behaves in a
this differentiation is due to an increase in IL-a and similar manner to Staphylococcus aureus a-toxin to
PGE2 (245, 355). A. actinomycetemcomitans lipopoly- induce apoptosis or necrosis in a dose-dependent
saccharide has also been reported to support the manner (135, 136). At high concentrations it is
survival of osteoclasts by the up-regulation and pro- suggested that the toxin binds non-speci®cally to cell
duction of IL-1a (355). Kawai et al. (146) have shown membranes forming large pores allowing the rapid
that A. actinomycetemcomitans lipopolysaccharide in¯ux of Ca2‡ and loss of ATP resulting in necrosis,
injected into the gingiva of rats could induce period- while at low concentrations the toxin is thought to
ontal bone resorption but only after the transfer of bind to speci®c cell surface proteins on susceptible
antigen-speci®c Th1 clone cells, but not after the cells and form small diameter pores allowing the
transfer of Th2 clone cells. These data support earlier uncontrolled in¯ux of Na‡ and the activation of
studies which showed that an A. actinomycetemco- apoptosis. Thus the A. actinomycetemcomitans leuko-
mitans-speci®c Th2 clone (A3) when adoptively toxin is a potent cytotoxic antigen able to kill immune
transferred into euthymic rats protected against cells, resulting in the dysregulation of the host
periodontal bone loss induced by oral infection with immune response and the release of a variety of
A. actinomycetemcomitans (53, 393). enzymes and reactive molecules from phagocytic cells
that may result in tissue damage and further in¯am-
mation. Although the majority of studies have
Leukotoxin
involved the ability of leukotoxin isolated from A. acti-
A. actinomycetemcomitans produces a 116 kDa imm- nomycetemcomitans serotype b to induce apoptosis,
unomodulating protein antigen, termed leukotoxin. Arakawa et al. (4) have reported that culture super-
The leukotoxin is a pore-forming protein and is a natants of serotypes b and a, but not c, induced apop-
member of the repeats-in-toxin (RTX) family of bac- tosis in HL-60 cells. Furthermore, scanning electron
terial toxins (182). In a study by Zambon et al. (403) micrographs con®rmed that serotypes b and a
®fty-®ve percent of localized aggressive periodontitis induced pore formation in HL-60 apoptotic cells. This
patients were shown to be infected with A. actino- ability of serotypes b and a, but not c, to induce
mycetemcomitans strains that produced a leukotoxin apoptosis is consistent with the serotype virulence
that was able to lyse human peripheral blood (b > a > c) in animal models (57).
polymorphonuclear leukocytes and HL-60 cells (a
promyelocytic cell line) in vitro. The leukotoxin is
Extracellular proteolytic enzymes
not released by A. actinomycetemcomitans but is
anchored by a C-terminal hydrophobic domain to Proteolytic activity in subgingival plaque, in particu-
the cellular outer membrane or membranous vesi- lar trypsin-like proteolytic activity, has been highly
cles (16, 181). A number of studies have shown that correlated with clinical measurements of periodon-
the killing of T-cells, neutrophils, natural killer (NK) titis (198, 268, 304). Culture supenatants of A. acti-
cells and polymorphonuclear leukocytes by leuko- nomycetemcomitans Y4 (serotype b) have been
toxin is by the formation of pores in the cell mem- shown to have high trypsin-like proteolytic activity
brane resulting in osmotic lysis and necrosis (132, (369). Wang et al. (369) puri®ed a 50-kDa trypsin-like
214, 312, 318, 340, 341). Unlike other members of protease from the culture supernatant of A. actino-
the RTX family, which kill a wide variety of cells, mycetemcomitans Y4 that hydrolysed the synthetic
A. actinomycetemcomitans leukotoxin is speci®c for substrates Na-benzoyl-dl-arginine p-nitroanilide
cells that express b2-integrin and lymphocyte func- and Na-benzoyl-dl-lysine p-nitroanilide as well as
tional antigen 1 (LFA-1). The binding of the leuko- the protein substrates, ®brinogen and collagen type
toxin to LFA-1, which is expressed on all immune I. Protease inhibition assays indicated that this

105
O'Brien-Simpson et al.

enzyme is a serine- or metallo-protease. Further- speculated, therefore that the long term infection
more, the tyrpsin-like protease reduced the prolifera- associated with chronic in¯ammatory diseases, such
tion of human gingival epithelial cells and cell as periodontitis, results in the constant exposure of
surface levels of ®bronectin (370). Gazi et al. (81) microbial heat shock antigens to the immune sys-
have reported that cell sonicates of A. actinomyce- tem, which may promote autoimmune disease (14,
temcomitans NCTC 9710 (serotype c) had no Arg-X 406).
activity and only a weak cysteine/serine protease Surface-associated material from A. actinomyce-
activity with alanine-X and lysine-X speci®c activity. temcomitans has been shown to induce bone resorp-
This difference in activity may be attributable to the tion in a murine calvarial assay (219). In a later study
different serotypes used in these studies. As by Kirby et al. (159) the potent osteolytic active com-
increased proteolytic activity of other periodontal ponent in the surface-associated material was iden-
pathogens, especially P. gingivalis, has been linked ti®ed as GroEL, as a monoclonal antibody to E. coli
to enhanced virulence, the greater proteolytic activ- GroEL inhibited A. actinomycetemcomitans-induced
ity, especially trypsin-like activity, of A. actinomyce- bone resorption. E. coli GroEL stimulates the recruit-
temcomitans serotype b compared with serotype c is ment and activation of osteoclasts in a dose depen-
consistent therefore with the greater virulence of this dent manner (286). These data therefore, suggest that
serotype. As well as degrading major components of A. actinomycetemcomitans GroEL may also recruit
connective tissue, proteolytic enzymes in A. actino- and activate osteoclasts. However, unlike E. coli
mycetemcomitans culture supernatants have been GroEL or A. actinomycetemcomitans LPS, where
reported to degrade IgG (all four subclasses), serum bone resorption activity is inhibited by indomethacin
(but not secretory) IgA and IgM but not IgD or IgE and dexamethasone, suggesting the mode of action is
(92) in vitro. Thus the proteolytic activity of A. actino- mediated by prostaglandins and IL-1 (126, 286),
mycetemcomitans may be an important virulence A. actinomycetemcomitans GroEL-induced bone
factor involved in dysregulation of the host's immune resorption is not inhibited by indomethacin or IL-1
response. receptor antagonist (219). As well as activating osteo-
clasts, A. actinomycetemcomitans GroEL induces
epithelial cell proliferation at low concentrations
GroEL heat shock protein
(0.4±1.0 mg/mL) and is cytotoxic at higher concen-
A. actinomycetemcomitans produces an antigenic trations (90, 265).
64 kDa GroEL protein that is equally expressed on
the cell surface of all serotypes a±e (238, 265). A.
Fimbriae
actinomycetemcomitans GroEL has 95% DNA
sequence identity with E. coli GroEL (159) and 57% A. actinomycetemcomitans produces highly anti-
and 42% sequence identity with M. tuberculosis heat genic (296) bundle-forming ®mbriae that are 5 nm
shock protein (hsp) 65 and human hsp60, respec- in diameter and several mm in length (289). Using
tively (238). In a study by Koga et al. (164) antibodies liquid chromatography±electrospray ionization mass
directed to A. actinomycetemcomitans GroEL were spectrometry (LC±MS) Inoue et al. (125) identi®ed
detected in 50% of sera taken from aggressive period- three ®mbrial sub-unit (®mbrillin) molecular mass
ontitis patients whereas no antibodies directed to species (6,226, 6,366, and 6,513 Da) which were
GroEL were detected in healthy patient sera. Antisera higher than the predicted molecular mass for the
raised against puri®ed A. actinomycetemcomitans ®mbrillin (5,118 Da) derived from the nucleotide
GroEL cross-react with E. coli GroEL, Mycobacterium sequence of the ®mbrillin gene ¯p (124). The discre-
bovis hsp 65, GroEL-like proteins from P. gingivalis pancies in the predicted and observed molecular
and T. forsythia and cross-react weakly with human masses of the ®mbrillin were attributed to post-
hsp60 (113, 338). Heat shock proteins have been translational glycosylations (125). The ®mbriae are
suggested to be associated with the aetiology and suggested to play an important role in colonization
pathogenesis of both experimental and naturally as ®mbriated A. actinomycetemcomitans strains have
occurring autoimmune diseases such as juvenile greater af®nity for epithelial cells and saliva-coated
chronic arthritis, rheumatoid arthritis and athero- hydroxyapatite than non-®mbriated strains (289).
sclerosis (91, 382). Microbial heat shock proteins, Fimbriae may also have a role in A. actinomycetem-
despite the high homology between prokaryotic comitans invasion of epithelial cells by receptor-
and eukaryotic cells, are highly immunogenic and mediated endocytosis (77). Harano et al. (110) have
can lead to antimicrobial immunity. It has been immunized rabbits with synthetic oligopeptides

106
 Antigens of bacteria associated with periodontitis

representing a part of the ®mbrial amino acid subjects. Beikler et al. (15) monitored the antibody
sequence. This immunization produced high titer subclass responses to a 110 kDa A. actinomycetem-
antibodies that recognized puri®ed ®mbrial protein. comitans outer membrane protein in periodontitis
Antisera raised to one of the ®mbrial peptides with patients 24 months after receiving full mouth scaling
the sequence VAVFYSNNGFIANLQSKFFSL was and/or antimicrobial therapy followed by supportive
reported to inhibit the attachment of ®mbriated periodontal therapy. Patients, habouring A. actinomy-
A. actinomycetemcomitans to saliva-coated hydro- cetemcomitans intraorally, with a high IgG4 response to
xyapatite beads, buccal epithelial cells and ®broblasts the 110 kDa protein had signi®cantly increased tooth
(110). These data along with earlier ®ndings that survival rates. No signi®cant association with clinical
®mbrial-speci®c high titre antibodies are associated treatment outcome was found for antigen-speci®c IgA,
with clearance of A. actinomycetemcomitans from IgG1, IgG2 or IgG3 serum levels, indicating that a serum
periodontal pockets in periodontitis patients (296) IgG4 antibody response to the 110 kDa protein may
indicate that a ®mbrial-based vaccine may be effective have a protective effect in A. actinomycetemcomitans-
in preventing A. actinomycetemcomitans infection. associated periodontitis (15).
Outer membrane and secreted proteins from
A. actinomycetemcomitans have also been reported
Other antigenic outer membrane
to be major inducers of in¯ammatory cytokines. Tani
proteins
et al. (348) isolated an extracellular 37 kDa antigenic
A number of studies (15, 56, 58, 60, 229, 372, 386) glycoprotein from A. actinomycetemcomitans Y4
have used patient antisera to probe cell sonicates or (serotype b) and evaluated its ability to induce the
outer membrane protein preparations of A. actino- release of cytokines from murine macrophages. The
mycetemcomitans in Western blots in order to iden- 37-kDa glycoprotein was found to induce strong
tify major antigenic proteins. Watanabe et al. (372) TNF-a, IL-1b, IL-6 responses and a weak granulocyte
found that a whole cell sonicate probed with sera macrophage conlony stimulating factor (GM-CSF)
from localized aggressive periodontitis patients response in macrophages (292). The IL-1b response
contained 13 major immunoreactive bands with a induced by the 37 kDa glycoprotein was similar to
molecular mass range of 14±78 kDa. The same inves- that induced by lipopolysaccharide whereas the IL-6
tigators showed than an outer membrane prepara- response induced by the glycoprotein was 8-fold
tion contained four main antigens of 19, 24, 35 and higher than that induced by lipopolysaccharide. As
67 kDa, that were immunoreactive with sera from well as inducing in¯ammatory cytokines that
these patients (372). Wilson and Hamilton (386) have enhance the innate immune response, A. actinomy-
compared the IgG subclass distribution in sera from cetemcomitans produces protein antigens that
localized aggressive periodontitis patients to 29.0 reduce cytokine production from T-helper cells and
and 16.6 kDa A. actinomycetemcomitans outer mem- thus potentially dysregulate the adaptive immune
brane proteins. For both proteins the IgG subclass response. Kurita-Ochiai and Ochiai (175) isolated a
distribution was IgG2 > IgG1 > IgG3. No IgG4 anti- 14-kDa protein from the supernatant of sonicated
bodies to the 29 or 16.6 kDa proteins were detected. A. actinomycetemcomitans Y4 which suppressed the
Ebersole et al. (56) have shown that A. actinomyce- proliferation of concanavalin A stimulated T-cells,
temcomitans has several other major outer mem- furthermore it was found to reduce the production
brane antigenic proteins with molecular masses of of Th1 cytokines IL-2 and IFN-g as well as Th2 cyto-
110, 90, 80, 58, 48 and 31 kDa. Furthermore, it was kines IL-4 and IL-5. Anti-human IL-10 antibody used
reported that the IgG3 subclass antibody in sera from in Western blot analysis to probe A. actinomycetem-
localized aggressive periodontitis patients bound to a comitans cell sonicates recognized a 65-kDa protein
wider range of A. actinomycetemcomitans outer and this protein was suggested to be capable of dys-
membrane proteins than IgG1 or IgG2. However, regulating the host immune response by binding to
an analysis of the frequency of patient sera that the IL-10 receptor as an agonist or antagonist (142).
bound to antigenic outer membrane proteins A. actinomycetemcomitans also produces other
showed that the IgG subclass distribution was immunosuppressive factors (78, 385). An 8-kDa anti-
IgG2 > IgG3 > IgG1 >> IgG4. Serum IgG antibody gen has been shown to suppress proliferation of
binding to 29 (383), 64 (238), 75 (296) and to 28, ®broblasts, monocytes and osteoblasts (218, 381) and
38, 54, 58, 75, and 90 kDa (60) outer membrane pro- 150 and 65 kDa proteins suppress the proliferation of
teins from periodontitis patients has been shown to human ®broblasts (141, 309). In a study by Shenker
be signi®cantly higher than for sera from healthy et al. (311) both T-cell and B-cell proliferation and

107
O'Brien-Simpson et al.

immunoglobulin production, but not T-cell IL-2 tors, nitric oxide, TNF-a and IL-1 from murine
release, were inhibited by a 60-kDa protein antigen macrophages (290). In this study the outer mem-
of A. actinomycetemcomitans. In a latter study by brane lipids and lipoproteins also induced the acti-
these investigators (313) the inhibition of T-cell and vation of the in¯ammatory cytokines, nitric oxide
B-cell activation was suggested to be the result of and TNF-a from lipopolysaccharide-unresponsive
the 60-kDa protein activating and stimulating differ- macrophages (290). Furthermore, polymixin B treat-
entiation of a T-cell subset that are CD4‡CD8‡ dual ment of the outer membrane lipid, but not the lipo-
positive. These T-suppressor cells were able to inhibit protein preparations, inhibited the activation of
the proliferation of mitogen stimulated T-cells (313). in¯ammatory cytokines in the lipopolysaccharide-
responsive and unresponsive macrophages (290).
Gopalsami et al. (89) have reported that the outer
Treponema denticola antigens membrane lipid of T. denticola was able to induce
bone resorption as measured by a 45Ca-release assay
of the radii and ulnae of 19-day-old fetal rats. How-
Lipopolysaccharide
ever, the addition of parathyroid hormone or pros-
T. denticola is a gram-negative, motile, asaccharoly- taglandin E2, which are known to enhance
tic, anaerobic spirochete with typical helical mor- lipopolysaccharide-induced bone resorption from
phology. Unlike lipopolysaccharide of other Gram- other gram-negative bacteria, had no additive effect.
negative bacteria the outer membrane sheath of These data suggest that the outer membrane lipids of
T. denticola lacks 3-deoxy-D-manno-octulosonic T. denticola induce bone resorption and activation of
acids, heptoses and b-hydroxy fatty acids and the in¯ammatory cytokines by a separate mechanism to
cytoplasmic membrane and outer membrane sheath lipopolysaccharide.
have similar fatty acyl chain composition (anteiso-
pentadecanoic acid, palmitic acid and iso-palmitic
Major sheath protein
acid) indicating that the membrane ¯uidity of
T. denticola is similar to the membranes containing A major surface protein and antigen of T. denticola is
lipoteichoic acid of gram-positive bacteria (302). the major sheath protein (Msp), which has been
Furthermore the membrane anchor consists of phos- reported to have different molecular masses in dif-
pholipid- and glycerolipid-like structures containing ferent T. denticola strains. The protein exists as a
two fatty acids (structurally similar to lipoteichoic 53-kDa protein in strains ATCC 35405 and 33520
acid) and not six as in lipid A of typical lipopolysac- and a 64-kDa protein in strain OKTI (359, 376). In
charide. However, T. denticola lipid does form nor- its native form the Msp aggregates to form oligo-
mal bilayer membranes with a distinct phase mers with reported molecular masses of 116±200
transition typical of lipopolysaccharide, but the (376) and 130±300 kDa (101). Although earlier stu-
phase transition of the outer sheath lipid is broader dies indicated that Msp formed a hexagonal array
and the temperature lower (228C) than for the phase structure in the outer sheath (215, 376), Caimano
transition of lipopolysaccharide (35.58C). These data et al. (24) has more recently suggested that Msp is
indicate that the ultrastructure of the outer mem- predominately a periplasmic protein that has lim-
brane sheath of T. denticola is similar to the outer ited surface exposure. This limited surface exposure
membrane of other gram-negative bacteria (389), may explain the differences in the antigenicity
however, the lipid composition of the outer sheath between strains ATCC 35405 and 33520 observed
is similar to lipoteichoic acid of the cell surface of by Umemoto et al. (359), even though there is
gram-positive bacteria (302). Although there are dif- 91% amino acid identity between the Msp proteins
ferences in the outer membrane of T. denticola and from both strains. Fenno et al. (72) suggested that
other gram-negative bacteria the outer membrane the surface exposed region of Msp is amino acids
lipids are reported to be highly antigenic as sera from 204±271. This 68 residue sequence is different in
periodontitis patients recognized lipid bands with both strains and therefore may help explain the
molecular masses 8±14 kDa in Western blot analysis difference in the reported antigenicity of the two
which were not recognized by sera from healthy strains. Kokeguchi et al. (166) have also shown that
individuals (401). Puri®ed outer membrane lipid rabbit antisera raised against the 53-kDa Msp pro-
and enriched outer membrane lipoprotein prepara- tein puri®ed from strain ATCC 35405 did not recog-
tions of T. denticola have been reported to induce, in nize the 53-kDa Msp protein from strain ATCC
a dose dependent manner, the in¯ammatory media- 33520 in a Western blot analysis of whole cell

108
 Antigens of bacteria associated with periodontitis

sonicates. As the proteins are 91% identical then

Extracellular proteolytic enzymes
this result further suggests that the 204±271 amino
acid sequence is immunodominant. The 53-kDa Important extracellular protein antigens of T. denti-
Msp protein binds ®brinogen, ®bronectin and lami- cola are its proteolytic enzymes, reviewed in Potempa
nin and thus may have an important role in binding et al. (276). The prolyl-phenylalanine speci®c, chymo-
of T. denticola (see below) to host cells (71, 101, trypsin-like, serine proteinase (357) known as dentili-
360). The Msp protein also aggregates with the chy- sin (127) or trepolisin (12), is the best characterized
motrypsin-like protease complex of T. denticola and T. denticola protease. This enzyme occurs on the cell
may play a role in binding to epithelial cells (358). surface (97) and has been puri®ed as a complex of three
The 64-kDa Msp protein from strain OKTI has also proteins, a 72-kDa proteinase encoded by the prtP gene
been shown to have an important role in the adhe- and two smaller proteins that have been reported in a
sion of T. denticola to gingival ®broblasts (375, 376). number of studies with different molecular masses,
As well as binding to host cell matrix proteins the possibly due to variable proteolytic processing. The
53-kDa Msp protein is toxic to HeLa cells, epithelial masses of the two smaller proteins have been reported
cells and erythrocytes (70, 73, 216). Although the as 38 and 43 kDa (127), 32 and 39 kDa (212) or 23 and
precise mechanism by which the 53-kDa Msp pro- 27 kDa (357). Puri®ed dentilisin from strain ATCC
tein induces cell death is not known, it has been 35405 has been shown to hydrolyse ®brinogen, trans-
shown to alter the membrane potential of arti®cial ferrin, gelatin, serum albumin, laminin, collagen
membranes and increase the membrane permeabil- type IV, IgG and IgA in vitro (97, 212, 357). The protei-
ity of HeLa cells, indicating that it may form a pore nase has also been shown to degrade bradykinin, sub-
in the host cell membrane resulting in unrestricted stance P and angiotensin I and II (212) as well as host
in¯ux of ions and eventual cell death (61, 216). The protease inhibitors a1-antitrypsin, antichymotrypsin,
ability of the Msp protein to bind to host matrix a2-macroglobulin, antithrombin III, antiplasmin and
proteins and form pores may, in part, explain the cystatin C (96). Dentilisin has also been suggested to
cytopathic effects of T. denticola on gingival ®bro- play a role in the degradation of pro-IL1b into its bioac-
blasts (8, 375) epithelial cells (154, 216, 358) lym- tive forms and thus stimulate the in¯ammatory
phocytes (310) and erythrocytes (70, 73, 93). The response (13). The ability to degrade host matrix pro-
53-kDa Msp protein also enhances the in¯amma- teins, in¯ammatory and protease regulatory proteins
tory response by triggering degranulation of poly- and peptides may contribute to the uncontrolled
morphonuclear cells and the speci®c release of degradation of periodontal tissues and enhance dis-
collagenase, gelatinase and matrix metalloprotei- ease progression. Ishihara et al. (128) have constructed
nases MMP-8 and MMP-9 (47). a dentilisin de®cient T. denticola isogenic mutant
based on strain ATCC 35405 by insertionally inactivat-
ing the prtP gene. They found that the dentilisin-de®-
Flagellum
cient mutant induced signi®cantly smaller lesions than
A major distinction between T. denticola and the the parent strain in the murine lesion model, indicating
other three periodontal pathogens discussed here is that dentilisin plays a signi®cant role in virulence.
that T. denticola is motile. The axial helical-¯agella T. denticola has been shown to bind to the ground
and the helically-shaped cell cylinder of T. denticola, substance glycosaminoglycan, hyaluronate via the
like all spirochetes, result in a cork screw-like loco- dentilisin complex (102). The dentilisin complex has
motion, which aids movement in highly viscous also been shown to bind and induce apoptosis in
environments like gingival crevicular ¯uid and the epithelial cells (73, 358). The prolyl-speci®c proteases,
penetration of gingival epithelial and connective proline iminopeptidase, prolyl oligopeptidase and
tissue (156, 161, 274, 293). The ¯agellum of T. denti- prolyl dipeptidyl peptidase that are believed to be
cola is composed of a basal body, rod, hook and a localized on the cell surface of T. denticola are
®lament. The ¯agellar ®lament consists of three core suggested to dysregulate the in¯ammatory response
proteins FlaB1 (35 kDa), FlaB2 (35 kDa) and FlaB3 by degrading vasoactive peptides, hormones and
(34 kDa) and a major sheath protein FlaA (38 kDa) neuropeptides (210, 211, 213).
(205, 294). Although each of the ®lament proteins are The presence of T. denticola in subgingival plaque
antigenic (37), FlaA and FlaB3 have been shown to be samples has been correlated with trypsin-like proteo-
the immunodominant antigens (237, 361). FlaA has lytic activity, which in turn has been correlated with
been suggested to have a role in adherence of T. den- clinical parameters of periodontitis (268, 304). Tryp-
ticola to host cells as it binds ®bronectin (360). sin-like activity has also been localized to the cell

109
O'Brien-Simpson et al.

surface of T. denticola (221). Fenno et al. (74) have T. forsythia or P. gingivalis cross-reacted in an ELISA,
isolated a trypsin-like enzyme encoded by the opdB Western and dot blot analyses to epitopes in lipid A
gene from T. denticola strain 35405 that cleaves the and core lipopolysaccharide carbohydrates, indicat-
synthetic substrate N-a-benzoyl-DL-arginine-2- ing antigenic and potentially structural and chemical
naphthylamide and natural peptides containing Arg similarity. Firoozkoohi et al. (75) analysed T. for-
and Lys residues. Trypsin-like activity has been sythia lipopolysaccharide using tricine±SDS±PAGE
found not to be responsible for the hydrolysis of a and showed that the lipopolysaccharide was a typical
variety of proteins including ®brinogen (208, 209, S-form exhibiting long ladders with multiple steps. The
211, 260), however, trypsin-like activity, and not same authors also showed that the O-polysaccharide
chymotrypsin-like activity, of T. denticola strain side chains of T. forsythia lipopolysaccharide are of
35405 was found to be important for the degradation similar length to those of Bacteroides species (75).
of human serum albumin (98).

Extracellular proteolytic enzymes

Other antigenic proteins
Several studies have shown that T. forsythia produces
A putative hemolysin, a 46-kDa antigenic protein cell surface proteolytic enzymes. These include the
named cystalysin is suggested to play a central role in trypsin-like serine proteases with molecular masses
iron acquisition by T. denticola as it is able to agglu- 70 and 81 kDa (95) and a 48-kDa hemolytic cysteine±
tinate and lyse erythrocytes and cause the oxidation protease designated PrtH (297). T. forsythia trypsin-
and sulphuration of hemoglobin (36, 171, 176). like activity, together with trypsin-like activities of
Although the T. denticola outer membrane anti- T. denticola and P. gingivalis in subgingival plaque
genic proteome is yet to be fully characterized, samples have been correlated with clinical para-
Kesavalu et al. (153) have reported that immunization meters of periodontitis (304). Interestingly, unlike
of mice with formalin-killed T. denticola cells induced P. gingivalis and T. denticola, T. forsythia has been
a strong antibody response to the majority of outer reported to be unable to proteolytically degrade a
membrane proteins as analysed by Western blot. variety of host protease inhibitors (96) and lactoferrin
(2). The trypsin-like activity has been suggested to
play a role in binding of T. forsythia to erythrocytes,
Tannerella forsythia (formerly polymorphonucleocytes and ®broblasts (230). Using
Bacteroides forsythus) antigens PCR Tan et al. (346) showed that T. forsythia occurred
in 91% of subgingival plaque samples from chronic
T. forsythia is a gram-negative, anaerobic, saccharo- periodontitis patients and that 85% of these were
lytic fusiform bacterium (349, 351). Very little is known colonized by T. forsythia with the prtH genotype,
about the antigens and virulence factors of T. forsythia whereas only 9% of subgingival plaque samples from
compared with the other major periodontal pathogens, healthy patients contained this genotype.
P. gingivalis, A. actinomycetemcomitans and T. denti-
cola. This has mainly been the result of the dif®culty in
Other antigenic outer membrane
reliably cultivating T. forsythia. However, the ®nding
proteins
that T. forsythia could be grown together with another
bacterium such as F. nucleatum or Streptococcus sanguis T. forsythia produces a protein that has been sug-
revealed that it requires exogenous N-acetylmuramic gested to induce apoptosis especially in lymphocytes
acid for growth. Inclusion of N-acetylmuramic acid in by the formation of membrane pores (4). A GroEL-
conventional media now allows routine cultivation of like protein with a molecular mass of 58 kDa is
T. forsythia in batch culture (392). also produced by T. forsythia (287) which may have
important implications for the immune response as
discussed in the section on A. actinomycetemcomi-
Lipopolysaccharide
tans antigens. Using periodontitis patient sera two
Very little is known about the structure and chemical antigenic proteins of masses 200 and 210 kDa were
composition of T. forsythia lipopolysaccharide, identi®ed in Western blot analysis as being common
however a study by Vasel et al. (365) indicated that to all of the T. forsythia strains tested (320).
T. forsythia lipopolysaccharide may be similar to A major surface antigen of T. forsythia is BspA, a
P. gingivalis lipopolysaccharide. Antibodies from 98-kDa protein (308). Sera from chronic periodontitis
rabbits immunized with lipopolysaccharide from patients, but not from healthy individuals, recog-

110
 Antigens of bacteria associated with periodontitis

nized recombinant BspA in a Western blot analysis

Table 1. Serotype classi®cations of P. gingivalis
(308). BspA contains 14 leucine-rich repeat (LRR)
motifs that are beleived to be involved in protein± Strain Capsular Fimbrial Outer
protein or lipid±protein interactions (162). Recombi- serotypea serotypec membrane
serotyped
nant BspA, containing the 14 LRR motifs was shown
to inhibit T. forsythia binding to ®brinogen and W50 K1 V C
®bronectin and a mutant strain de®cient in BspA W83 K1 V C
had reduced ability to bind to these proteins (120, HG184 K2 ± ±
308). BspA also has been shown to stimulate proin-
A7A1-28 (ATCC 53977) K3 II B
¯ammatory cytokine production in THP-1 mono-
nuclear cells via interaction with CD14 and Toll- ATCC 49417 K4 II ±
like receptor 4 (104). HG1690 K5 ± ±
HG1691 K6 ± ±

Porphyromonas gingivalis antigens

b
381 K-neg I A(D)c
2561 (ATCC 33277) K-negb I A
P. gingivalis is a gram-negative, non-spore forming, FAY19m-1 ± III ±
non-motile, asaccharolytic, obligate anaerobic coc-
9-14K1 ± IV ±
cobacillus. A number of studies have attempted to
a
classify P. gingivalis into serotypic groups based on b
(179, 388).
P. gingivalis strain 381 and strain 2561 (ATCC 33277) are reported as K
antisera binding to either the capsule, ®mbriae or antigen negative (K-neg) capsule deficient (31, 67, 179, 388).
c
(189, 234).
outer membrane antigens of P. gingivalis (Table 1) d
e
(55).
Nagata et al. (232) have suggested that 381 may represent a separate
(55, 179, 189, 388). outer membrane serotype (serotype D).

Capsule
One major surface antigen of P. gingivalis is the cap- negative strains induced only a localized abscess
sule, a polysaccharide heteropolymer up to 15 nm (180). P. gingivalis capsule also inhibits the attach-
thick, that surrounds the outer membrane (390). ment of periodontal ligament ®broblasts to the tooth
Based on the antigenicity of the polysaccharide K root surface and long-term exposure to capsular-
antigens of the capsule, six K antigen serotypes have polysaccharide alters the properties of the tooth root
been reported (Table 1) (179, 388). P. gingivalis surface decreasing the ability of ®broblasts to attach
strains 381 and ATCC 33277 are considered K anti- (405). Encapsulated strains of P. gingivalis K1 (W50)
gen negative as they lack a capsular-speci®c anti- and K2 (HG184) are also more resistant to phago-
body response and strain 381 has no capsular- cytosis by polymorphonuclear cells than a K-nega-
polysaccharide layer as determined by electron tive strain 381 (335). The decreased ability to activate
microscopy and phase contrast analysis (31, 67, the alternative complement pathway and the
179). The capsule of P. gingivalis strain A7A1-28 increase in cellular hydrophilicity due to the capsu-
has been shown to be immunochemically distinct lar polysaccharide are proposed mechanisms by
from lipopolysaccharide, to contain galactosamine, which encapsulated P. gingivalis is resistant to
glucosamine, galactosaminuronic acid and glucose phagocytosis (335, 364, 388).
and to have a high amino acid content (301). Califano
et al. (26) have shown that antisera from chronic
Lipopolysaccharide
periodontitis patients recognized all six K-antigen
serotypes and that the IgG isotype elicited was Several studies have suggested that lipopolysacchar-
almost exclusively IgG2. Furthermore, serotypes K1 ide of P. gingivalis is an important antigen in the host
(strain W50), K2 (strain HG184) and K6 (strain immune response in periodontitis. Lipopolysacchar-
HG1691) were recognized more frequently. The cap- ide has been shown to be a major P. gingivalis anti-
sule of P. gingivalis is suggested to be an important gen recognized by sera from periodontitis patients
virulence factor as strains representing the six K- with the antibody being of the IgG2 subclass (68, 199,
antigen positive serotypes have been shown to be 300). P. gingivalis lipopolysaccharide lacks heptose
highly invasive causing a spreading ulcerative lesion and b-hydroxydecanoic acid typically found in clas-
in the murine lesion model, whereas K-antigen sical enterobacterial lipopolysaccharide (163, 233),

111
O'Brien-Simpson et al.

but exhibits typical S-form and smooth lipopolysac- ide receptor, but via Toll-like receptors (TLRs) as
charide chemotype (21). The 2-keto-3-deoxyoctonate antibodies to TLR-4 inhibited IL-6 expression (336).
present in P. gingivalis lipopolysaccharide, unlike Furthermore, human gingival ®broblasts with high
other gram-negative lipopolysaccharide, is phos- surface levels of TLR-4 expressed higher levels of
phorylated (at position C7 or C8) and is structurally IL-1 and IL-6 in response to P. gingivalis lipopoly-
essential for the attachment of the polysaccharide saccharide stimulation compared with ®broblasts
component (79, 172). Furthermore, the lipid A com- expressing low levels of TLR-4 on their surface (371).
ponent of P. gingivalis lipopolysaccharide contains Wendell and Stein (380) have shown that a combina-
high amounts of iso-branched chain and hydroxy tion of P. gingivalis lipopolysaccharide and nicotine
fatty acids and no straight alkane chains (C14, had a synergistic effect upregulating the expression
C18:O) (117, 173). The chemical structure of lipid A of IL-6 and IL-8 in human gingival ®broblasts. These
of strain 381 is a glucosamine-b-(1-6) disaccharide- authors suggested that a combination of bacterial
1-monophosphate acylated with 3-hydroxy-15- lipopolysaccharide and nicotine from smoking, a
methylhexadecanoic acid and 3-hexadecanoyloxy- known risk factor for periodontitis (249), provides
15-methylhexadecanoic acid at positions 2 and 2' an environment conducive for the stimulation of
respectively (253). The chemical structure of the O- pro-in¯ammatory cytokines and the potentiation of
polysaccharide of lipopolysaccharide from P. gingi- disease. As well as stimulating pro-in¯ammatory
valis has been studied for strains ATCC 53978 (69) cytokines from ®broblasts P. gingivalis lipopolysac-
and W50 (266) each showing unique and different charide also induces the secretion of nitric oxide,
structural and chemical compositions. The chemical TNF-a, PGE2 and IL-1 from human and murine
structure of the O-polysaccharide for strain ATCC macrophages (21, 306, 347). P. gingivalis lipopoly-
53978 is a tetrasaccharide repeat unit of !4)-a-L- saccharide was found to stimulate these pro-in¯am-
Rhap-(1!3)-a-d-Galp-(1!3)-a-d-Galp-(1!4)-GlcNAc matory cytokines to a similar level in macrophages
(69) and strain W50 O-polysaccharide is a tetrasac- isolated from E. coli lipopolysaccharide-responsive
charide of !6)-a-d-Glcp (1!4)-a-L-Rhap-(1!3)-b- (CD14‡) mice (murine strain C3H/HeN) and E. coli
d-GalNAc-(1!3)-a-D-Galp where the a-rhamnose is lipopolysaccharide-unresponsive (CD14 ) mice (mur-
non-stoichiometrically phosphorylated with mono- ine strain C3H/eJ) (21, 347). This is again consistent
phosphoethanolamine (266). These differences and with P. gingivalis lipopolysaccharide binding to TLRs
unusual modi®cations of the O-antigen of P. gingi- and not CD14 and in fact macrophages have recently
valis lipopolysaccharide between strains and with been shown to be stimulated via TLR-2 to secrete
classical gram-negative lipopolysaccharide are likely pro-in¯ammatory cytokines by P. gingivalis lipopo-
to have important implications for the immunogenic lysaccharide (116). Neutrophils exposed to P. gingi-
properties of P. gingivalis lipopolysaccharide and dis- valis lipopolysaccharide have increased expression of
ease pathogenesis. P. gingivalis lipopolysaccharide the transcription factors NF-kB and AP-1 and
induces a variety of cytokines from a number of enhanced expression of the pro-in¯ammatory cyto-
different cells. Human gingival ®broblasts secrete kine IL-8 (333). The stimulation of neutrophil IL-8
the pro-in¯ammatory cytokines IL-1b, IL-6 and IL- secretion was found to be partly inhibited (44%) with
8 in response to incubation with P. gingivalis lipo- anti-CD14 antibodies indicating that P. gingivalis
polysaccharide (342, 345). Reddi et al. (285) have lipopolysaccharide stimulates pro-in¯ammatory
shown that the lipid A component is more biologi- cytokine expression in neutrophils by at least two
cally active than whole lipopolysaccharide in indu- pathways. Jotwani et al. (137) have shown that den-
cing IL-6 in human gingival ®broblasts and a dritic cells pulsed with P. gingivalis lipopolysacchar-
macrophage cell line. The induction and stimulation ide undergo maturation, up-regulate accessory
of IL-6 expression and secretion in human gingival molecules and secrete IL-1b, PGE2, IL-10 and IL-12.
®broblasts by P. gingivalis lipopolysaccharide is These dendritic cells were shown to have a reduced
enhanced by the simultaneous induction of IL-1b ability to stimulate T helper (Th) cell proliferation in
(394). P. gingivalis lipopolysaccharide also stimulates vitro with a reduced release of IFN-g and prevented a
the secretion of a higher amount of IL-6 from gingival Th-polarization in vivo. Similarly, Pulendran et al.
®broblasts isolated from periodontitis patients than (280) have reported the stimulation of counter reg-
those isolated from healthy patients (149). ulatory cytokines from dendritic cells and Th-cells,
The mechanism by which P. gingivalis lipopoly- where dendritic cells pulsed with P. gingivalis lipo-
saccharide stimulates IL-6 secretion in ®broblasts polysaccharide up-regulated co-stimulatory mole-
is suggested to be not via CD14, the lipopolysacchar- cules and produced in¯ammatory cytokines IL-6

112
 Antigens of bacteria associated with periodontitis

and TNF-a. Whereas P. gingivalis lipopolysaccharide

Fimbriae
stimulated Th cells secretion of signi®cantly higher
levels of Th-2 cytokines IL-13, IL-5 and the Th reg- The ®mbriae of P. gingivalis are highly antigenic with
ulatory cytokine IL-10. This ability of P. gingivalis chronic and aggressive periodontitis patients exhibit-
lipopolysaccharide to stimulate counter regulatory ing high serum IgA and IgG antibody responses, pre-
cytokines from non-myeloid, myeloid and lymphoid dominantly an IgG3 isotype, to ®mbrial antigen
cells using different cellular receptors may aid in the compared with healthy subjects (256, 399). The ®m-
dysregulation of the host immune response and thus briae are curly, single-stranded ®laments, 5 nm in
exacerbate disease. diameter and are composed of ®mbrillin (®mbrial
In vitro murine calvarial and murine periodontitis sub-unit), a 43-kDa protein encoded by the ®mA
models have been used to show that P. gingivalis gene (46, 398). Currently ®ve P. gingivalis ®mbrial
lipopolysaccharide is a potent inducer of bone types (types I±V) have been described based on their
resorption by the stimulation of local IL-1a and IL- antigenicity, (Table 1) (189, 234). P. gingivalis ®m-
1b (123, 243). Miyata et al. (224) have shown that the briae are thought to be important for the adhesion
interaction of P. gingivalis lipopolysaccharide with of the bacterium to host tissues (129, 190, 307). A
CD14 on osteoclasts induces the expression of IL- number of studies have reported that the ®mbriae
1b and IL-6, which in turn activates the cells and of P. gingivalis strains 381 and ATCC 33277 (type I
initiates bone resorption. IL-1 receptor and TNF ®mbriae) are essential for binding to, and invasion of,
receptor signaling of osteoclasts after stimulation at oral epithelial cells and ®broblasts (50, 129, 246, 374,
high, but not low, doses of P. gingivalis lipopolysac- 377). Furthermore, P. gingivalis ®mbrial mutants of
charide has also been suggested to be important 381 or ATCC 33277 have been shown to have reduced
mechanisms for bone resorption (35). Furthermore, or no ability to bind and invade epithelial or ®bro-
P. gingivalis lipopolysaccharide also promotes differ- blast cells (105, 246, 377). Contradictory evidence
entiation in vitro of osteoclasts isolated from lipo- exists for the adherence and invasion of epithelial
polysaccharide-responsive and lipopolysaccharide- and ®broblast cells of sparsely ®mbriated P. gingiva-
unresponsive mice indicating that P. gingivalis can lis strains W50 and W83 (type V ®mbriae). Two stu-
activate osteoclasts via a separate pathway than dies suggested that P. gingivalis W50 and W83 do not
CD14 signaling (131). Although high molecular mass adhere to or invade epithelial and ®broblast cells (50,
P. gingivalis lipopolysaccharide induces more bone 374). However, Sandros et al. (299) have shown that
loss than low molecular mass material (222), P. gingivalis strain W50 does adhere to epithelial
Bom-van Noorloos et al. (19) have shown that other cells, albeit weaker than strain 381, and that it was
heat-sensitive P. gingivalis components have a able to invade epithelial cells to the same level as
greater role in activating osteoclasts and inducing the highly ®mbriated strains with invasion being
bone resorption. These results may indicate the pre- characterized by receptor-mediated endocytosis.
sence of lipopolysaccharide-associated proteins in Furthermore, Nakagawa et al. (236) have shown that
these lipopolysaccharide preparations. P. gingivalis microspheres conjugated with recombinant ®mbrilin
lipopolysaccharide and lipid A delay apoptosis and proteins from each of the ®ve ®mbrial types adhered
reduce the production of fragmented DNA in neu- and invaded human gingival ®broblasts and epithe-
trophils and polymorphonuclear cells (115, 277). In lial cells. These authors further showed that P. gingi-
contrast, Isogai et al. have suggested that P. gingiva- valis strains with type II ®mbriae had a greater ability
lis lipopolysaccharide induces apoptosis in T-cells, to invade epithelial cells (236). In this study by
B-cells expressing immunoglobulin, Ia and CD5 Nakagawa et al. (236) the adherence and invasion
and inactivating lymphocytes in lymph nodes, spleen of epithelial cells was suggested to be mediated by
and thymus (130). Taken together these data suggest ®mbrial interaction with the surface receptor a a5b1
that P. gingivalis lipopolysaccharide induces bone integrin. Sojar et al. (329) have reported that P. gin-
resorption and stimulates the release of pro-in¯am- givalis ®mbriae bind to cytokeratin on the surface of
matory cytokines from non-myeloid and myeloid epithelial cells and suggested that this interaction
cells and inhibits apoptosis of these cells important may be important for colonization of the oral mucosa
in the innate immune response, whereas the adap- by P. gingivalis. Active invasion of endothelial cells
tive immune response is dysregulated by stimulating by P. gingivalis mediated via ®mbriae stimulates the
the release of counter regulatory cytokines from dif- expression of the surface cell adhesion molecules
ferent lymphoid cells and enhancing apoptosis in ICAM-1, VCAM-1, P- and E-selectins, thus recruiting
activated lymphoid cells. leukocytes to the site and enhancing the in¯ammatory

113
O'Brien-Simpson et al.

response (155). P. gingivalis ®mbriae have also been sized overlapping peptides representing the 43 kDa
shown to stimulate the release of IL-1a, IL-1b TNF-a, ®mbrillin (Fim A) sub-unit of P. gingivalis and con-
neutrophil chemotactic factor KC, and MCP-1 from jugated them to thyroglobulin (22). These overlap-
macrophages and ®broblasts and TNF-a, IL-6 and ping peptide±thyroglobulin conjugates were then
IL-8 from polymorphonuclear cells (106±108, 231, used to immunize wistar rats and the resultant anti-
257, 258). The induction of IL-1b and TNF-a in sera used to probe either native or denatured ®m-
murine macrophages by P. gingivalis is thought to brillin. It was found that only antisera to peptide 99±
be stimulated by ®mbrial binding to b2 integrin (344). 110 recognized native ®mbrillin indicating that this
Similarly THP-1 cells (a human macrophage cell line) epitope is accessible to antibody binding in the
incubated with recombinant FimA produced pro- native protein. T-cell epitopes for P. gingivalis Fim
in¯ammatory cytokines via CD14-TLR4 and b2 integ- A ®mbrillin have also been identi®ed using the
rin receptor signaling (104). In this same study, PEPSCAN methodology. Ogawa et al. (254) identi®ed
P. gingivalis minor ®mbriae, distinct from FimA, six T-cell epitopes by evaluating the T-cell prolifera-
induced the release of pro-in¯ammatory cytokines tive response of P. gingivalis ®mbrillin primed
from THP-1 cells via TLR-2 receptor signaling. T-cells against 67 overlapping 10-mer synthetic pep-
Furthermore, ®mbrial stimulation of IL-8 and NF- tides representing the amino acid sequence of the
kB secretion from epithelial cells has been suggested ®mbrillin. Furthermore, three of the identi®ed T-cell
to be via TLR-2 receptor signaling (6). The stimulated epitopes were found to be adjacent to immuno-
release of IL-1 and GM-CSF by P. gingivalis ®mbriae dominant B-cell epitopes. Synthetic peptides incor-
has been shown to stimulate bone resorption in a porating a T-cell epitope and the corresponding
murine embryonic calvarial bone resorption assay by adjacent B-cell epitope were synthesized and
activating osteoclasts (147). induced a protective immune response in guinea-
Immunization with the P. gingivalis 43 kDa ®m- pigs against lesion development after subcutaneous
brillin and ®mbrial peptides protects against P. gin- challenge with P. gingivalis strain 381 (254). Deslaur-
givalis strain 381 (type I ®mbriae) induced bone loss iers et al. (45) have also identi®ed P. gingivalis ®m-
in the rat periodontitis model (29, 64, 254, 255). Nasal brillin T-cell epitopes using overlapping 20-mer
vaccination of rats with ®mbriae with the mucosal peptides. One 20-mer peptide (PgF-P8) also con-
adjuvant cholera toxin induced a ®mbriae-speci®c tained a B-cell epitope and mice immunized with
IgA response and these antibodies inhibited the PgF-P8 had a 60% survival rate when challenged with
attachment of P. gingivalis strain 381 to oral epithe- P. gingivalis ATCC 33277 compared to 25% survival
lial cells and reduced the cytokine in¯ammatory in control animals. Fimbrial protein and peptides are
response of epithelial cells to P. gingivalis (396). therefore capable of protecting against P. gingivalis-
Baker et al. (10) have shown that in a murine period- induced lesions and periodontal bone loss in animal
ontitis model, P. gingivalis 381, despite colonizing models. However, a vaccine based on one ®mbrial
the gingival crevice, did not cause periodontal bone type may be strain speci®c and hence ineffective
loss whereas sparsely ®mbriated strains W50 (type V against other P. gingivalis strains of different ®mbrial
®mbriae), A7A1-28 (type II ®mbriae) and DPG3 a types.
de®mbriated mutant strain of 381 did colonize and
induce periodontal bone loss. Fan et al. (67) have
Extracellular proteolytic enzymes
shown that antibodies to type I ®mbriae opsonized
and enhanced the phagocytosis of P. gingivalis strain P. gingivalis produces a variety of proteinases and
381 and ATCC 33277 but not P. gingivalis strains their ability to degrade host proteins and modulate
from the other ®mbrial types (types II±V). These the immune response has been recently reviewed by
results suggest that P. gingivalis strains with type I Potempa et al. (276), therefore only those protei-
®mbriae may be less virulent, as the type I ®mbriae nases found to be highly antigenic will be reviewed
assist the host response in preventing establishment here. The Arg-X and Lys-X speci®c extracellular
of the bacterium. cysteine proteinases and their associated adhesins
Ogawa et al. (259) synthesized 102 overlapping are highly antigenic (235, 251). Three genes encode
peptides representing another ®mbrial protein (Pg- the major extracellular Arg-X and Lys-X speci®c
II) of P. gingivalis and found that sera from patients cysteine proteinases of P. gingivalis. These are rgpA,
with chronic periodontitis reacted with 7 epitopes rgpB and kgp (42). rgpA encodes an Arg-X proteinase
whereas sera from healthy patients did not recognize with a C-terminal extension of adhesins. rgpB
any of the overlapping peptides. Brant et al. synthe- encodes an Arg-X proteinase without the C-terminal

114
 Antigens of bacteria associated with periodontitis

Arg R 1 E 2 13 5 4 5 4 21 E2
RgpA
polyprotein
RgpA45 RgpA44 RgpA15 RgpA27
100aa RgpA17
Lys K 1 E 2 135 4 5 421 42
Kgp
polyprotein
Kgp48 Kgp39 Kgp15 Kgp13
Kgp14 Kgp20
Proteinase Active Site Peptides
R- PAS1R RgpA45(199-219) FNGGISLANYTGHGSETAWGT
K- PAS1R Kgp48(204-224) LNTGVSFANYTAHGSETAWAD
Adhesin Binding Motif Peptides
1- ABM1 RgpA45(436-462) PYQPVSNLTATTQGQKVTLKWDAPSTK
2- ABM2 Kgp39(658-686) SYTYTVYRDGTKIKEGLTATTFEEDGVAA
3- ABM3 RgpA44(712-744) VTLKWDAPNGTPNPNPNPNPNPNPGTTTLSESF
4- ABM4 Kgp14(1068-1087) WIERTVDLPAGTKYVAFRHY
5- ABM5 RgpA15(927-942) PAEWTTIDADGDGQGW
E- EP1 RgpA44(604-628) DDYVFEAGKKYKFLMKKMGSGDGTE
Control Peptide
RgpA27(1205-1236) ANEAKVVLAADNVWGDNTGYQFLLDADHNTFG

Fig. 1. Schematic representation of the RgpA and Kgp

polyproteins of P. gingivalis W50 showing the location
and amino acid sequence of the synthesized peptides.
Similarly hatched areas at other locations correspond to
repeat sequences (sometimes with substitutions) of the
same motif. ( ) indicates the processing site for the pro-
teinase and adhesin domains and ( ) indicates the puta- Fig. 2. 2-D gel electrophoresis of P. gingivalis W50 outer
tive lipopolysaccharide attachment site (250). membrane prepared from heme-limited chemostat-grown
cells (367).
extension and kgp encodes a Lys-X proteinase with a
C-terminal extension of adhesins that are sequence RgpA/B and Kgp proteinases are abundant proteins
related to the rgpA adhesins (42). The proteins (367). Analysis of whole P. gingivalis W50 cell Arg-X
encoded by rgpA and kgp of strain W50 form sur- and Lys-X proteolytic activity (252) together with the
face-associated complexes of non-covalently asso- speci®c activities of the puri®ed enzymes (322) indi-
ciated proteinases and adhesins, designated the cates that the RgpA/B and Kgp proteinases represent
RgpA-Kgp proteinase±adhesin complexes, formerly approximately 10% w/w of the total amount of
the PrtR-PrtK proteinase±adhesin complexes (17, protein produced by P. gingivalis W50 under heme
367). The RgpA±Kgp complexes of P. gingivalis strain excess growth conditions and corresponds to approx-
W50 are characterized by a 45 kDa Arg-X-speci®c imately 100,000 copies of the proteinase molecules
proteinase (RgpA45) associated with four sequence- on the surface of each cell. These data con®rm the 2D
related adhesins, RgpA44, RgpA15, RgpA17 and PAGE results that the RgpA/B and Kgp proteinases
RgpA27 all encoded by rgpA (Fig. 1). The RgpA-Kgp represent the majority of surface exposed protein on
complexes are also characterized by a 48 kDa Lys-X- P. gingivalis W50 (Fig. 3).
speci®c proteinase (Kgp48) associated with three In a recent study the RgpA-Kgp proteinase±adhe-
sequence-related adhesins Kgp39, Kgp15 and Kgp44 sin complexes were shown to be highly antigenic as
all encoded by kgp (17, 322, 323). Using 2D PAGE
and peptide mass ®ngerprinting the Kgp44 adhesin
has recently been shown to be proteolytically pro-
cessed into three subunits of 14 kDa, 13 kDa and
potentially a lipopolysaccharide-associated 20 kDa
C-terminal subunit (Fig. 1) (367). The extracellular
Arg-X-speci®c cysteine proteinase RgpB (322) is not
found associated with adhesins as the rgpB gene does
not encode adhesins nor an adhesin binding motif
(see below) that is present in the RgpA and Kgp
catalytic domains (322). RgpB has been isolated as
both a 70±80 kDa membrane-associated protein
with lipopolysaccharide-covalently attached at the
C-terminus (367) and as a discrete C-terminally-
truncated 50-kDa protein from the culture superna-
tant (284, 322). 2D-PAGE analysis of P. gingivalis Fig. 3. Schematic presentation of the outer membrane of
W50 outer membranes (Fig. 2) revealed that the P. gingivalis.

115
O'Brien-Simpson et al.

92% of patients with chronic periodontitis had a ®ve sequences have been identi®ed that are
serum IgG response to these complexes (251). repeated, with substitutions, throughout both the
Furthermore, an increase in IgG antibody response RgpA and Kgp adhesins (Fig. 1) (250). Based on
to RgpA-Kgp was shown to be positively correlated sequence similarity with known adhesins and data
with clinical parameters of disease and the percen- showing involvement in RgpA-Kgp complex aggrega-
tage of periodontal pockets infected with P. gingiva- tion (316, 322) these repeat sequences have been
lis. However, the IgG isotype antibody response to designated adhesin binding motifs (ABM1-5; Fig. 1).
RgpA-Kgp was found to be predominantly IgG2 and As well as the ABM peptides a major antigenic deter-
IgG4 and the levels of these two IgG isotypes differed minant (EP1) has been identi®ed using murine and
with disease severity. As severity of disease increased human protective sera to probe overlapping syn-
there was a corresponding increase in the IgG2 anti- thetic peptides representing the RgpA27 and Kgp39
body response (a Th-1 cytokine induced antibody adhesins by PEPSCAN analysis (Fig. 1) (250, 251).
(160)) and a decrease in the IgG4 response (a Th-2 Synthetic peptides corresponding to EP1 and three
cytokine induced antibody (80, 204)). Using Western of the ®ve ABM sequences (ABM 1±3) when used as
blot analysis, the major immunodominant proteins immunogens gave protection against challenge with
in the RgpA-Kgp complexes were found not to be the P. gingivalis in the murine lesion model (250). Unlike
proteinases but the non-covalently bound adhesins, the protective ABM1-3 peptides, ABM 4 and ABM 5
in particular RgpA44 and the potential vaccine can- and a control peptide RgpA27 (1205±1236) (Fig. 1)
didates Kgp39 and RgpA27, which were recognized that did not protect against challenge with P. gingi-
by sera from patients with low levels of disease (251). valis, induced low titre, cross-reactive antibodies to
Sera from mice immunized with the RgpA-Kgp com- the RgpA-Kgp complexes. Furthermore, using the
plexes were found to be immunoreactive with the PEPSCAN methodology the predominant IgG sub-
adhesins RgpA44; Kgp39 and RgpA27 (250). Further- class antibody that bound to the protective epitopes
more, it has been shown that these same adhesins in both studies was shown to be Th2 cytokine
were recognized by sera from rats protected against induced IgG antibodies, IgG4 in humans and IgG1
P. gingivalis-induced periodontal bone loss after in mice (250, 251). The three synthetic ABM peptides
being immunized with the RgpA-Kgp complexes (ABM 1±3) and EP1 that conferred protection exist
(282). This has been corroborated in a recent study within a 120 residue span in the RgpA44 and Kgp39
by Gibson and Genco (85); who showed that mice adhesins of the RgpA-Kgp complexes. Booth et al.
immunized with the RgpA proteinase and its asso- (20) have demonstrated that passive immunization
ciated adhesins protected against P. gingivalis of humans with a monoclonal antibody (Mab
induced periodontal bone loss whereas the RgpB 61BG1.3) restricted colonization of P. gingivalis in
proteinase did not result in a protective immune the oral cavity. The binding site of this Mab was
response. Although immunization with the protei- mapped using recombinant RgpA44 to residues
nase catalytic domain may lead to a non-protective 907±931 (148) which is within the 120 residue span
immune response, synthetic peptides representing mentioned above. Part of the sequence recognized
the proteinase active sites of RgpA/B and Kgp pro- by Mab 61BG1.3 included the sequence PVXNLT
teinases (PAS1R and PAS1K; Fig. 1) conjugated to which is found in ABM1. The same sequence has also
diphtheria toxoid have been shown to protect mice been identi®ed using phage display epitope mapping
against P. gingivalis induced lesions (250). Further- techniques as a haemagglutinin-associated motif
more, a synthetic peptide representing the N-termi- (316). This 120 residue antigenic and functionally-
nus of the RgpA/B proteinase catalytic domains has relevant segment therefore has potential as a recom-
also been reported to protect mice against P. gingi- binant protein vaccine against P. gingivalis-induced
valis challenge in the murine chamber model (82). periodontitis.
These data suggest that the proteinase catalytic The RgpA/B and Kgp proteinases of P. gingivalis
domains have non-protective immunodominant are believed to be major virulence factors for the
antigenic determinants that de¯ect the immune bacterium (9, 150, 239, 282, 326). The proteolytic
response away from protective epitopes on the pro- activity of these enzymes has been shown to degrade
teinase, and that a directed immune response against a variety of host proteins in vitro e.g. ®brinogen,
these speci®c, proteinase sequences can lead to pro- ®bronectin, laminin and collagen types III, IV and
tection. V (186, 188, 275, 276, 303). However, the major degra-
By analysing the amino acid sequence of the RgpA dation of host tissue is unlikely to involve these pro-
and Kgp proteinases and their associated adhesins teinases directly but rather indirectly by activation of

116
 Antigens of bacteria associated with periodontitis

latent host matrix metalloproteinases (330). P. gingi- and binding to matrix proteins relative to the wild-
valis proteinases have also been reported to degrade type 381 strain. Further, a triple mutant based on
plasma host defense/regulatory proteinase inhibi- strain ATCC 33277 lacking RpgA, RgpB and Kgp does
tors; a-antitrypsin, a2-macroglobulin, antichymo- not agglutinate erythrocytes, bind to hemoglobin or
trypsin, antithrombin III and antiplasmin (27, 96). grow in de®ned media containing bovine serum
In¯ammation at the site of infection may be albumin as the sole carbon/energy source (314). Pro-
enhanced by the proteinases as the complement pro- teinase activity and the adhesin domains of the
tein C5 has been reported to be hydrolysed in vitro RgpA-Kgp complexes have been suggested to play a
releasing the biologically active C5a fragment (334). role in the binding and invasion of oral epithelial
Cell lysis mediated by complement however, may be cells by P. gingivalis (33, 185, 353). The RgpA and
avoided as the proteinases have been reported to Kgp proteinases have also been reported to induce
degrade the other complement proteins (334, 387). apoptotic cell death in human gingival ®broblasts
The P. gingivalis proteinases have also been shown and epithelial cells (7, 34, 368). Recently, the contri-
to activate and dysregulate the kallikrein/kinin and bution of RgpA, RgpB and Kgp to the virulence of
coagulation/®brinolysis pathways resulting in the P. gingivalis W50 was shown in the murine lesion
uncontrolled production of the pro-in¯ammatory model (252) using isogenic mutants lacking each
mediators bradykinin and thrombin, that may proteinase (1, 284). All of the isogenic mutants had
further contribute to the in¯ammatory immune signi®cantly reduced virulence in the murine lesion
response (reviewed in Potempa et al. (276)). Phago- model compared with the wild type strain and that
cytic and other functions of neutrophils recruited to the order in which the gene products contributed to
the in¯amed site may also be impaired, as the pro- virulence was Kgp >> RgpB  RgpA (252). These
teinases degrade cell surface receptors in vitro (133, studies therefore suggest that RgpA, RgpB and Kgp
139). Furthermore, the RgpA/B proteinases have are critical for the virulence of P. gingivalis.
been reported to degrade CD14 (a pattern recogni-
tion receptor for bacterial components) on the sur-
face of monocytes inhibiting lipopolysaccharide GroEL heat shock protein
activation (332) and on human gingival ®broblasts
leading to a down regulation of lipopolysaccharide The P. gingivalis heat shock protein GroEL is highly
induced IL-8 production (339). The RgpA/B protei- antigenic and anti-P. gingivalis GroEL antibodies
nases have also been reported to activate protease- have been detected in gingival tissue extracts (337).
activated receptors (PARS) on platelets resulting in Furthermore, Tabeta et al. (337) have shown that
platelet aggregation (201) and on oral epithelial cells there was a higher proportion of sera from period-
inducing IL-6 secretion (200). The host immune ontitis patients that recognized P. gingivalis GroEL
response may be further dysregulated by the protei- than from healthy subjects. Epitope mapping (PEPS-
nases as they have been reported to degrade IgG, IgA, CAN) of the P. gingivalis 60 kDa GroEL using sera
IgM, IgD and IgE (76, 92). Furthermore, spontaneous from chronic periodontitis patients showed that the
P. gingivalis mutants with reduced Arg-X and Lys-X majority of patient sera tested recognized only four
proteinase activity and wild-type cells treated with an immunodominant epitopes (206). Furthermore, all of
Arg- and Lys-speci®c protease inhibitor (Na-p-Tosyl- the patient sera tested recognized the peptide
l-Lysine Chloro-methyl Ketone, TLCK) have been sequence MQFDRGYISP that has high sequence
reported to be avirulent in animal models (150). A homology to an epitope in human heat shock protein
beige mutant of P. gingivalis W50/BEl, which has 60 (hsp60) with the sequence MKFDRGYISP.
reduced Lys-X and Arg-X proteinase activity, is Although only two periodontitis patients had antibo-
reported to be avirulent in animal models. However, dies directed to human hsp60, Maeda et al. (206)
this mutant also has reduced gelatinase, collagenase, suggested that an antibody response to P. gingivalis
and dipeptidylaminopeptidase activity, as well as GroEL may lead to human hsp60 cross reactivity and
reduced haemagglutinin activity, ®mbriation and an autoinimune reaction. Unlike human hsp60,
extracellular vesicle production. The change in these P. gingivalis GroEL was unable to stimulate periph-
factors may also contribute to this mutant's reduced eral blood monocytes to produce Th1 or Th2 cyto-
virulence (38, 217, 305). Tokuda et al. (353, 354) have kines (395) or to stimulate the release of pro-
reported that two constructed isogenic mutants of in¯ammatory cytokines from macrophages (356).
P. gingivalis strain 381 lacking either RgpA or RgpB Although these data show that P. gingivalis GroEL
exhibited reduced aggregation, haemagglutination is antigenic and that there is antibody cross-reactivity

117
O'Brien-Simpson et al.

between hsp60 and P. gingivalis GroEL epitopes the antigenic outer membrane protein with a molecular
lack of cross-reactivity of P. gingivalis GroEL with mass of 30 kDa (initially designated Pga30 (111), now
hsp60-speci®c monocytes, macrophages and T-cells IhtB (43)) from P. gingivalis W50 has been suggested
is not consisitent with a role of the P. gingivalis pro- to be part of an iron transport system (43). IhtB has
tein in triggering an autoimmune reaction (356, 395). signi®cant sequence similarity to a S. typhimurium
cobalt chelatase (CbiK) and has recently been shown
to exhibit chelatase activity (288) thus it has been
Haemagglutinins
suggested that the peripheral outer membrane che-
The cell surface haemagglutinating adhesin, HA-Ag2, latase of P. gingivalis may remove iron from heme
is a surface antigen of P. gingivalis as all of the sera prior to cell uptake (43). IhtB was strongly immunor-
from 8 chronic periodontitis patients reacted with eactive with sera from a patient who haboured
two HA-Ag2 proteins with masses of 43 and 49 kDa P. gingivalis in subgingival plaque but exhibited no
(44). Furthermore HA-Ag2 was a common antigen to clinical signs of periodontitis (111).
all the P. gingivalis strains tested (48, 227). The two A number of studies have investigated the anti-
HA-Ag2 proteins have been described as having genic pro®le of the outer membrane of P. gingivalis
masses of either 33 and 38 kDa (227) or 43 and and have identi®ed a range of immunodominant
49 kDa (44). These discrepancies have been attribu- proteins by their molecular mass. Watanabe et al.
ted to different extraction techniques (228). The 33 (373) probed P. gingivalis W50 whole cell sonicate
and 38 kDa HA-Ag2 proteins were extracted using an and outer membrane protein preparations in a 1D
EDTA containing buffer whereas the 43 and 49 kDa Western blot analysis using sera from 103 chronic
forms were isolated from extracellular vesicles. Spe- periodontitis patients and identi®ed six major anti-
ci®c antibodies to HA-Ag2 have been shown to inhi- genic outer membrane proteins; 75, 57, 51, 46, 35 and
bit haemagglutination and binding of P. gingivalis to 19 kDa and three major antigens in the sonicate
epithelial cells (28, 49, 315). A number of studies have material (46, 27, and 14 kDa). The 46 kDa antigen
identi®ed four P. gingivalis genes that encode hae- was suggested to be the predominant immunoreac-
magglutinin proteins HagA, B, C, and D (191, 192, tive protein with sera from patients with severe
278, 279). Research into these four Hag proteins has periodontitis however the protein was not identi®ed.
focused on HagA and B with particular interest in Curtis et al. (40) inactivated the RgpA/B and Kgp
HagB as a potential vaccine candidate. Katz et al. proteinases of P. gingivalis prior to extraction and
(145) have shown that antibodies against recombi- isolation of outer membranes and reported three
nant HagB recognize a 50 kDa protein (HagB) in immunodominant antigens with mass 115, 55 and
P. gingivalis strains; ATCC 33277, 381, A7A1±28 and 40 kDa. Sera from chronic periodontitis patients
W50 indicating that HagB is a common P. gingivalis have been shown to recognize three proteins from
antigen. Mice intragastrically inoculated with an P. gingivalis strain 381 with masses 82, 57 and
avirulent strain of Salmonella typhimurium expres- 44 kDa, whereas sera from aggressive periodontitis
sing the hagB gene mounted both a systemic and patients recognized lower molecular mass proteins
mucosal antibody response and that response could of 44, 27, 25 and 18 kDa (187). By investigating the
be boosted indicating that a memory T-cell and B- antigenic cross-reactivity of the four P. gingivalis
cell response was induced (51, 52, 165). Furthermore, outer membrane serotypes (serotypes A±D; ATCC
rats immunized subcutaneously with recombinant 33277 (A), A7A1±28 (B), W50 (C) and 381 (D)) using
HagB were protected against periodontal bone loss antibodies to intact cells, four proteins with masses
induced by P. gingivalis strain ATCC 33277 (145). 140, 130, 37 and 28 kDa common to all four serotypes
were identi®ed (66). Furthermore, all of the antibo-
dies raised to each of the serotype-speci®c strains
Antigenic outer membrane proteins
were able to opsonize and enhance phagocytosis
Curtis and co-workers, have identi®ed a 55 kDa by polymorphonuclear cells of all the other sero-
major antigenic outer membrane protein of P. gingi- types. Whole cell sonicates of P. gingivalis 381 were
valis W50 that was immunoreactive with all of the 26 probed in a Western blot analysis using sera from 77
periodontitis patients sera tested but was weakly or periodontitis patients and a 53 kDa (Ag53) was
non-reactive with sera from healthy individuals. This reported as the immunodominant P. gingivalis anti-
protein, designated RagB and a co-transcribed gene gen recognized by sera from 53 of the periodontitis
product RagA, have been suggested to play a role in patients (174). The B-cell epitopes of Ag53 have been
active transport in P. gingivalis (40, 41, 223). Another characterized using the PEPSCAN approach. In a

118
 Antigens of bacteria associated with periodontitis

study by Oyaiza et al. (263) sera from chronic period- 13±42 kDa higher than that calculated from their
ontitis patients recognized three major epitopes of gene sequences and as the proteins were reactive
Ag53 and the predominant antibody subclass that with an lipopolysaccharide-reactive monoclonal
recognized these epitopes was the Th1 cytokine- antibody, it was concluded that all these proteins
induce antibody IgG2. Ohyama et al. (261) have also were lipopolysaccharide-modi®ed (367). The PMF
identi®ed human HLA-DRB1 speci®c Ag53 T-cell data supported this conclusion as no C-terminal
epitopes. Short-term T-cell lines derived from patient peptides were identi®ed, which is consistent with
peripheral blood mononuclear cells (PBMC) were the presence of a modi®cation. A Western blot from
stimulated with 45 overlapping peptides represent- a 2D-gel was also performed using an identical OM
ing the amino acid sequence of Ag53, 12 T-cell epi- sample to that used to identify the proteins listed in
topes were identi®ed with one peptide epitope Fig. 2 and Table 2, and by using antisera to P. gingi-
immunodominant. These peptides were reactive valis whole cells. A number of resolved spots (or
with T-cells from aggressive periodontitis patients trains of spots) were produced from which several
but not with T-cells from healthy individuals. major antigens could be positively assigned to
Although Ag53 has been shown to be highly antigenic Omp40/41, RagB and the adhesins of RgpA and
it is not expressed by all P. gingivalis strains and is Kgp. An area of the blot corresponding to a pI range
speci®c for serotypes A and D (119), thus it may have of approximately 4±6 and extending from a molecu-
limited use as P. gingivalis vaccine. lar mass of about 45 kDa to the top of the blot was
The majority of the outer membrane protein anti- almost completely covered with stain. As lipopoly-
gens (14±115 kDa) in these early studies were not saccharide is known to be highly antigenic, the stain-
identi®ed by sequence analysis, highlighting the ing most likely corresponded to the presence of the
need for a comprehensive analysis of P. gingivalis lipopolysaccharide-modi®ed proteins listed above
outer membrane antigens. (RgpA27, RgpB, P27, P59) together with less abun-
dant lipopolysaccharide-modi®ed proteins which
were detected now due to the increased sensitivity
Outer membrane proteome analysis
of Western blots compared to protein staining
The recently released P. gingivalis genomic sequence methods.
has enabled the rapid identi®cation and characteri- The surface-associated, lipopolysaccharide-attached
zation of surface proteins using 2D-PAGE and mass proteins of P. gingivalis therefore appear to be the
spectrometric techniques (Fig. 2) (366, 367). Thirty- most antigenic (Fig. 3). Lipopolysaccharide-attach-
®ve outer membrane proteins with masses ranging ment may serve not only to allow these proteins to
from 15±110 kDa, similar to those reported above, be localized at some distance from the cell surface,
have been identi®ed in P. gingivalis W50 using this but may also serve to protect the proteins from
approach (Tables 2 and 3). Peptide mass ®ngerprint- the host immune system. The integral OMPs and
ing (PMF) proved to be essential for the identi®cation TonB-linked receptors (Fig. 3) may be only weakly
of these proteins as up to 20 were N-terminally antigenic, possibly due to their long ¯exible extra-
blocked by the conversion of the N-terminal gluta- cellular loops that make binding to antibody entro-
mine to pyroglutamate or the fatty acid modi®cation pically unfavourable. The porin, Omp40/41, due to
of the N-terminal Cys residue of putative lipoproteins their shear abundance are dominant antigens that,
(367). The proteins identi®ed included many integral like their counterparts in other bacteria, may have
OMPs and TonB-linked receptors, putative OM lipo- vaccine potential (291).
proteins and putative peripheral proteins (Fig. 2).
The sequence-related cysteine proteinases, adhesins
and haemagglutinins encoded by the genes rgpA, Concluding remarks
rgpB, kgp and hagA, together with the newly identi-
®ed P27 and P59 were found to share signi®cant Each of the periodontal pathogens discussed in this
sequence similarity in their C-terminal 50 amino acid review produces an array of antigens that stimulate
residues to each other and to ®ve other proteins epithelial cells and ®broblasts to produce pro-
encoded by P. gingivalis genes that were not identi- in¯ammatory cytokines. Furthermore, these antigens
®ed in the study, presumably due to lack of abun- stimulate pro-in¯ammatory cells such as macro-
dance (367). As RgpB, P27, P59 and the C-terminal phages, monocytes and neutrophils and these in turn
domain of RgpA (RgpA27) migrated on 2D-gels as produce a wide variety of cytokines that maintain a
vertical streaks at an apparent molecular weight pro-in¯ammatory response and induce a Th1

119
O'Brien-Simpson et al.

Table 2. Identi®cation data for P. gingivalis W50 OM-associated proteins (except RgpA, Kgp and HagA)
Protein Mr Obs. (103) Mass calc. (kDa) Description (reference) Accession #
RgpB 71±98 55.6 Proteinase (322) AF007124
RagA 98 110 TOMR (109) AJ130872
RagB 52 54.4 OM Lipoprotein (109) AJ130872
T1r 76 76.7 TOMR (324) AF155223
IhtB 31 30.7 OM Lipoprotein (43) AAF03904
Omp40 43 40.4 Putative porin (366) PG0626
Omp41 43 41.3 Putative porin (366) PG0627

P15 15 15.6 OM Lipoprotein PG1177

P61 57 60.5 OM Lipoprotein PG0916

P90 86 90.4 TOMR PG1752

P92 91 92 TOMR PG0637

P93 86 92.6 TOMR PG1242
P49 49 48.7 TolC. PG1455
P27 43±64 26.9 Surface protein PG1570
P59 71±98 58.4 Surface protein PG1838

P20 21 19.9 Integral Omp PG1592

P22 24 22.0 Integral Omp PG1840

P23 27 23.0 Integral Omp PG0551

P26 26 25.6 Integral Omp PG0836

P30 30 30.2 Integral Omp PG1562

P40 43 40.2 Integral Omp PG0023

P64 57 64.1 Integral Omp PG1442

P51 50 51.4 Contains OmpA motif PG1893

P58 54 57.5 ± PG1424

The theoretical mass of each protein was calculated from its predicted mature form. The accession numbers containing the prefix `PG' are gene identification
numbers from the Oral Pathogen Sequence Databases obtained from [Link]

Proteins designated Pxx where xx is the calculated molecular mass.

(in¯ammatory) cytokine response. Although the possible explanation of why this may occur at the
induction and maintenance of an in¯ammatory site of infection is that the mixed response may, in
response by the immune system to pathogenic sub- part, re¯ect the dysregulation of the host defence in
gingival plaque bacteria is consistent with the period- periodontal lesions thus allowing bacteria at the site
ontal diseases being in¯ammatory diseases, a number of infection to evade an effective immune response.
of studies have shown that some antigens of the Proteolytic activity, in particular `trypsin-like' pro-
pathogenic bacteria also stimulate a Th2 cytokine teolytic activity of subgingival plaque has been
response which is known to be anti-in¯ammatory. positively correlated with clinical parameters of per-
This induction of pro-in¯ammatory and anti-in¯am- iodontitis (198, 268, 304). The four pathogens dis-
matory cytokines by antigens from periodontal cussed in this review that have been highly
pathogens is interesting and requires further inves- associated with periodontitis all produce extracellu-
tigation. However, the induction of Th1 and Th2 lar proteolytic activity including `trypsin-like' prote-
cytokines is not mutually exclusive and several inves- lytic activity. These enzymes are likely to be involved
tigators have found a mixed pro- and anti-in¯amma- directly or indirectly in destruction of host tissue and
tory cytokine response in periodontal tissue. A dysregulation of host defences by activating latent

120
 Antigens of bacteria associated with periodontitis

Table 3. Identi®cation data for processed domains of poor compliance of the patient. The use of antibio-
RgpA, Kgp and HagA tics, both systemic and local, is limited by the need
Processed domain Mr Obs./mass N-terminal for constant treatment to prevent reestablishment of
(residues numbered calc. (kDa) sequence§ the pathogen. The elucidation of the speci®c bacter-
from initial Met) ial aetiology of periodontitis suggests that the devel-
RgpA45 (228±688) 47/51 R-YVEEKQ opment of a speci®c treatment modality to target site
RgpA44 (720±1081) 
46/38 R-SGQAEIVL colonization or virulence of P. gingivalis, T. forsythia,

T. denticola and A. actinomycetemcomitans is now a
RgpA17 (1274±1404) 15/15 K-PQSVWIER
more rational approach to treat the disease. The sig-

RgpA27 (1432±1706) (43±64)/30 R-ANEAKVVL ni®cant reduction of periodontal disease progression
Kgp48 (229±710) 
50/53 R-DVYTDHGD in the non-human primate and rodent by immuniza-
 tion with either killed whole P. gingivalis cells or
Kgp39 (738±1099) 42/38 R-ANEAKVVL
y
P. gingivalis antigens suggests that vaccination may
Kgp 14 (1292±1424) 14/15 K-PQSVWIER
be an important adjunctive therapy to mechanical
y
HagA30 (366±625) 30/28 K-APAPYQER debridement in humans to prevent recolonization
HagA32 (820±1077), 32/28 K-PQSVWIER of periodontal pathogens. The development of a
(1272±1529)y multispecies vaccine that is able to target all four
HagA15 (1724±1856)y 15/15 K-PQSVWIER bacterial species that have been implicated in the

development of periodontitis may be more success-
RgpA15 (1139±1257) 18/13 R-ADFTETFE
Kgp15 (1157±1275)
ful than a vaccine against a single species. Vaccina-
HagA18 (685±803), tion may also have a therapeutic bene®t as although
(1137±1255), the bacteria may be more resistant to the adaptive
(1589±1707)y immune response when present in subgingival pla-
The identified processed domains of RgpA, Kgp and HagA are listed que as an undisturbed bio®lm, speci®c antibodies
together with their observed and calculated molecular mass, and their
observed N-terminal sequences (367). The amino acid sequences of RgpA
may still restrict the progression of disease by block-
and Kgp were obtained from Slakeski et al. (321) (Genbank accession ing the penetration of the major virulence factors
AAC18876) and Slakeski et al. (323) (Genbank accession U75366)
respectively. The sequence of HagA was obtained from the Oral Pathogen into the gingival tissues. As well as blocking these
Sequence Databases at [Link] where it has the gene
identification number PG1602.

virulence factors, the antibodies, if directed against
Nomenclature taken from Bhogal et al. (17) and O'Brien-Simpson et al.
(252). key epitopes involved in function, may neutralize
y
Nomenclature based on name of protein and Mr observed.
§
Hyphen indicates cleavage site. their action and also facilitate their removal through
opsonization and phagocytosis. Speci®c antibodies
may also neutralize key virulence factors associated
with acquisition of essential nutrients, thereby
host enzymes and degrading host matrix proteins, restricting proliferation within the plaque bio®lm.
cellular receptors, proteinase inhibitors, complement, Furthermore, speci®c antibodies of a certain subclass
antibodies and certain cytokines. They are also likely (IgA, IgG4 in humans) may reduce in¯ammation
to be involved in adherence to host cells and other associated with bacterial infections at mucosal sites.
bacteria and therefore in colonization of the gingival However, the development of a vaccine is dependent
crevice. The extracellular proteinases RgpA/B and Kgp on the identi®cation of bacterial antigens that are
of P. gingivalis have been the most widely investigated expressed in vivo and induce a protective response.
proteases of the four pathogens. These P. gingivalis Thus it is important to use a combined proteomic,
enzymes are essential for virulence in animal models. genomic and immunological strategy to identify bac-
Similarly an extracellular proteinase of T. denticola terial antigens of periodontal pathogens and evaluate
has been shown to be important for virulence in an their potential as vaccine candidates for the devel-
animal model of disease. T. forsythia and A. actino- opment of a multispecies vaccine for periodontitis as
mycetemcomitans proteinases are also likely to be an adjunct to current periodontal therapies.
important virulence factors and therefore require
further investigation.
The current treatment of periodontitis is non-spe-
References
ci®c and is centered on the removal of subgingival
plaque by mechanical debridement often involving 1. Aduse-Opoku J, Davies NN, Gallagher A, Hashim A, Evans
surgical procedures. This on-going therapy is costly, HE, Rangarajan M, Slaney JM, Curtis MA. Generation of
painful and has a variable prognosis due in part to lys-gingipain protease activity in Porphyromonas gingivalis

121
O'Brien-Simpson et al.

W50 is independent of Arg-gingipain protease activities. of Arg- and Lys-specific cysteine proteinases and adhesins.
Microbiology 2000: 146: 1933±1940. Microbiol 1997: 143: 2485±2495.
2. Alugupalli KR, Kalfas S. Degradation of lactoferrin by per- 18. Bird PS, Shakibaie F, Gemmell E, Polak B, Seymour GJ.
iodontitis-associated bacteria. FEMS Microbiol Lett 1996: Immune response to Bacteroides forsythus in a murine
145: 209±214. model. Oral Microbiol Immunol 2001: 16: 311±315.
3. Amano K, Nishihara T, Shibuya N, Noguchi T, Koga T. Im- 19. Bom-van Noorloos AA, van der Meer JW, van de Gevel JS,
munochemical and structural characterization of a serotype- Schepens E, van Steenbergen TJ, Burger EH. Bacteroides
specific polysaccharide antigen from Actinobacillus actino- gingivalis stimulates bone resorption via interleukin-1 pro-
mycetemcomitans Y4 (serotype b). Infect Immun 1989: 57: duction by mononuclear cells. The relative role for B. gin-
2942±2946. givalis endotoxin. J Clin Periodontol 1990: 17: 409±413.
4. Arakawa S, Nakajima T, Ishikura H, Ichinose S, Ishikawa I, 20. Booth V, Ashley FP, Lehner T. Passive immunization with
Tsuchida N. Novel apoptosis-inducing activity in Bacter- monoclonal antibodies against Porphyromonas gingivalis
oides forsythus: a comparative study with three serotypes of in patients with periodontitis. Infect Immun 1996: 64:
Actinobacillus actinomycetemcomitans. Infect Immun 422±427.
2000: 68: 4611±4615. 21. Bramanti TE, Wong GG, Weintraub ST, Holt SC. Chemical
5. Armitage GC. Development of a classification system for characterization and biologic properties of lipopolysac-
periodontal diseases and conditions. Ann Periodontol charide from Bacteroides gingivalis strains W50, W83,
1999: 4: 1±6. and ATCC 33277. Oral Microbiol Immunol 1989: 4:
6. Asai Y, Ohyama Y, Gen K, Ogawa T. Bacterial fimbriae and 183±192.
their peptides activate human gingival epithelial cells 22. Brant EE, Sojar HT, Sharma A, Bedi GS, Genco RJ, De
through Toll-like receptor 2. Infect Immun 2001: 69: Nardin E. Identification of linear antigenic sites on the
7387±7395. Porphyromonas gingivalis 43-kDa fimbrillin subunit. Oral
7. Baba A, Abe N, Kadowaki T, Nakanishi H, Ohishi M, Asao T, Microbiol Immunol 1995: 10: 146±150.
Yamamoto K. Arg-gingipain is responsible for the degrada- 23. Bumann D, Meyer TF, Jungblut PR. Proteome analysis of
tion of cell adhesion molecules of human gingival fibro- the common human pathogen Helicobacter pylori. Proteo-
blasts and their death induced by Porphyromonas mics 2001: 1: 473±479.
gingivalis. Biol Chem 2001: 382: 817±824. 24. Caimano MJ, Bourell KW, Bannister TD, Cox DL, Radolf JD.
8. Baehni PC, Song M, McCulloch CA, Ellen RP. Treponema The Treponema denticola major sheath protein is predo-
denticola induces actin rearrangement and detachment of minantly periplasmic and has only limited surface expo-
human gingival fibroblasts. Infect Immun 1992: 60: 3360± sure. Infect Immun 1999: 67: 4072±4083.
3368. 25. Califano JV, Schenkein HA, Tew JG. Immunodominant anti-
9. Baker PJ, Evans RT, Roopenian DC. Oral infection gen of Actinobacillus actinomycetemcomitans Y4 in high-
with Porphyromonas gingivalis and induced alveolar responder patients. Infect Immun 1989: 57: 1582±1589.
bone loss in immunocompetent and severe combined 26. Califano JV, Schifferle RE, Gunsolley JC, Best AM,
immunodeficient mice. Arch Oral Biol 1994: 39: 1035± Schenkein HA, Tew JG. Antibody reactive with Porphyro-
1040. monas gingivalis serotypes K1 in adult and generalized
10. Baker PJ, Dixon M, Evans RT, Roopenian DC. Heterogene- early-onset periodontitis. J Periodontol 1999: 70: 730±735.
ity of Porphyromonas gingivalis strains in the induction of 27. Carlsson J, Herrmann BF, Hofling JF, Sundqvist GK.
alveolar bone loss in mice. Oral Microbiol Immunol 2000: Degradation of the human proteinase inhibitors alpha-1-
15: 27±32. antitrypsin and alpha-2-macroglobulin by Bacteroides gin-
11. Barlow DJ, Edwards MS, Thornton JM. Continuous and givalis. Infect Immun 1984: 43: 644±648.
discontinuous protein antigenic determinants. Nature 28. Chandad F, Mouton C. Antigenic, structural, and func-
1986: 322: 747±748. tional relationships between fimbriae and the hemagglu-
12. Barrett A, Rawlings N, Woessner FE. Handbook on Protei- tinating adhesin HA-Ag2 of Porphyromonas gingivalis.
nases. Amsterdam: Academic Press, 1998. Infect Immun 1995: 63: 4755±4763.
13. Beausejour A, Deslauriers N, Grenier D. Activation of the 29. Chen PB, Davern LB, Schifferle R, Zambon J. Protective
interleukin-1 beta precursor by Treponema denticola: a immunization against experimental Bacteroides (Porphyro-
potential role in chronic inflammatory periodontal dis- monas) gingivalis infection. Infect Immun 1990: 58: 3394±
eases. Infect Immun 1997: 65: 3199±3202. 3400.
14. Beck JD, Offenbacher S, Williams R, Gibbs P, Garcia R. 30. Chen PB, Davern LB, Neiders ME, Reynolds HS, Zambon JJ.
Periodontitis: a risk factor for coronary heart disease? Analysis of in vitro lymphoproliferative responses and
Ann Periodontol 1998: 3: 127±141. antibody formation following subcutaneous injection of
15. Beikler T, Karch H, Ehmke B, Klaiber B, Flemmig TF. Pro- Actinobacillus actinomycetemcomitans and Wolinella recta
tective effect of serum antibodies against a 110-kilodalton in a murine model. Oral Microbiol Immunol 1991: 6: 12±16.
protein of Actinobacillus actinomycetemcomitans following 31. Chen HA, Weinberg A, Darveau RP, Engel D, Page RC.
periodontal therapy. Oral Microbiol Immunol 1999: 14: Immunodominant antigens of Porphyromonas gingivalis
281±287. in patients with rapidly progressive periodontitis. Oral Mi-
16. Berthold P, Forti D, Kieba IR, Rosenbloom J, Taichman NS, crobiol Immunol 1995: 10: 193±201.
Lally ET. Electron immunocytochemical localization of Ac- 32. Chen PB, Davern LB, Katz J, Eldridge JH, Michalek SM.
tinobacillus actinomycetemcomitans leukotoxin. Oral Mi- Host responses induced by co-infection with Porphyromo-
crobiol Immunol 1992: 7: 24±27. nas gingivalis and Actinobacillus actinomycetemcomitans
17. Bhogal PS, Slakeski N, Reynolds EC. A cell-associated pro- in a murine model. Oral Microbiol Immunol 1996: 11:
tein complex of Porphyromonas gingivalis W50 composed 274±281.

122
 Antigens of bacteria associated with periodontitis

33. Chen T, Nakayama K, Belliveau L, Duncan MJ. Porphyro- 49. Du L, Pellen-Mussi P, Chandad F, Mouton C, Bonnaure-
monas gingivalis gingipains and adhesion to epithelial Mallet M. Fimbriae and the hemagglutinating adhesin HA-
cells. Infect Immun 2001: 69: 3048±3056. Ag2 mediate adhesion of Porphyromonas gingivalis to
34. Chen Z, Casiano CA, Fletcher HM. Protease-active extra- epithelial cells. Infect Immun 1997: 65: 3875±3881.
cellular protein preparations from Porphyromonas gingi- 50. Duncan MJ, Nakao S, Skobe Z, Xie H. Interactions of Por-
valis W83 induce N-cadherin proteolysis, loss of cell phyromonas gingivalis with epithelial cells. Infect Immun
adhesion, and apoptosis in human epithelial cells. J 1993: 61: 2260±2265.
Periodontol 2001: 72: 641±650. 51. Dusek DM, Progulske-Fox A, Whitlock J, Brown TA. Isola-
35. Chiang CY, Kyritsis G, Graves DT, Amar S. Interleukin-1 tion and characterization of a cloned Porphyromonas gin-
and tumor necrosis factor activities partially account for givalis hemagglutinin from an avirulent strain of
calvarial bone resorption induced by local injection of Salmonella typhimurium. Infect Immun 1993: 61: 940±946.
lipopolysaccharide. Infect Immun 1999: 67: 4231±4236. 52. Dusek DM, Progulske-Fox A, Brown TA. Systemic and mu-
36. Chu L, Holt SC. Purification and characterization of a cosal immune responses in mice orally immunized with
45 kDa hemolysin from Treponema denticola ATCC avirulent Salmonella typhimurium expressing a cloned
35404. Microb Pathog 1994: 16: 197±212. Porphyromonas gingivalis hemagglutinin. Infect Immun
37. Cockayne A, Sanger R, Ivic A, Strugnell RA, MacDougall 1994: 62: 1652±1657.
JH, Russell RR, Penn CW. Antigenic and structural ana- 53. Eastcott JW, Yamashita K, Taubman MA, Harada Y, Smith
lysis of Treponema denticola. J Gen Microbiol 1989: 135: DJ. Adoptive transfer of cloned T helper cells ameliorates
3209±3218. periodontal disease in nude rats. Oral Microbiol Immunol
38. Collinson LM, Rangarajan M, Curtis MA. Altered expres- 1994: 9: 284±289.
sion and modification of proteases from an avirulent mu- 54. Ebersole JL. Systemic humoral immune responses in per-
tant of Porphyromonas gingivalis W50 (W50/BE1). iodontal disease. Crit Rev Oral Biol Med 1990: 1: 283±331.
Microbiology 1998: 144: 2487±2496. 55. Ebersole JL, Steffen MJ. Human antibody responses to
39. Consensus report. Periodontal diseases: pathogenesis and outer envelope antigens of Porphyromonas gingivalis sero-
microbial factors. Ann Periodontol 1996: 1: 926±932. types. J Periodontal Res 1995: 30: 1±14.
40. Curtis MA, Slaney JM, Carman RJ, Johnson NW. Identi- 56. Ebersole JL, Cappelli D, Sandoval MN, Steffen MJ. Antigen
fication of the major surface antigens of Porphyromonas specificity of serum antibody in A. actinomycetemcomi-
gingivalis using IgG antibody reactivity of periodontal tans-infected periodontitis patients. J Dent Res 1995: 74:
case-control serum. Oral Microbiol Immunol 1991: 6: 658±666.
321±326. 57. Ebersole JL, Kesavalu L, Schneider SL, Machen RL, Holt SC.
41. Curtis MA, Hanley SA, Aduse-Opoku J. The rag locus of Comparative virulence of periodontopathogens in a mouse
Porphyromonas gingivalis: a novel pathogenicity island. abscess model. Oral Dis 1995: 1: 115±128.
J Periodontal Res 1999: 34: 400±405. 58. Ebersole JL, Hall EE, Steffen MJ. Antigenic diversity in the
42. Curtis MA, Kuramitsu HK, Lantz M, Macrina FL, Nakayama periodontopathogen, Actinobacillus actinomycetemcomi-
K, Potempa J, Reynolds EC, Aduse-Opoku J. Molecular tans. Immunol Invest 1996: 25: 203±214.
genetics and nomenclature of proteases of Porphyromonas 59. Ebersole JL, Feuille F, Kesavalu L, Holt SC. Host modula-
gingivalis. J Periodontal Res 1999: 34: 464±472. tion of tissue destruction caused by periodontopathogens:
43. Dashper SG, Hendtlass A, Slakeski N, Jackson C, Cross KJ, effects on a mixed microbial infection composed of Por-
Brownfield L, Hamilton R, Barr I, Reynolds EC. Charac- phyromonas gingivalis and Fusobacterium nucleatum. Mi-
terization of a novel outer membrane hemin-binding crob Pathog 1997: 23: 23±32.
protein of Porphyromonas gingivalis. J Bacteriol 2000: 60. Ebersole JL, Steffen MJ, Cappelli D. Longitudinal human
182: 6456±6462. serum antibody responses to outer membrane antigens of
44. Deslauriers M, Mouton C. Immunoreactivity in humans of Actinobacillus actinomycetemcomitans. J Clin Periodontol
Bacteroides gingivalis hemagglutinating adhesin HA-Ag2. 1999: 26: 732±741.
Oral Microbiol Immunol 1990: 5: 302±304. 61. Egli C, Leung WK, Muller KH, Hancock RE, McBride BC.
45. Deslauriers M, Haque S, Flood PM. Identification of mur- Pore-forming properties of the major 53-kilodalton surface
ine protective epitopes on the Porphyromonas gingivalis antigen from the outer sheath of Treponema denticola.
fimbrillin molecule. Infect Immun 1996: 64: 434±440. Infect Immun 1993: 61: 1694±1699.
46. Dickinson DP, Kubiniec MA, Yoshimura F, Genco RJ. Mo- 62. Eke PI, Rotimi VO, Laughon BE. Coaggregation of black-
lecular cloning and sequencing of the gene encoding the pigmented Bacteroides species with other oral bacteria.
fimbrial subunit protein of Bacteroides gingivalis. J Bacter- J Med Microbiol 1989: 28: 1±4.
iol 1988: 170: 1658±1665. 63. Ellen RP, Schwarz-Faulkner S, Grove DA. Coaggregation
47. Ding Y, Uitto VJ, Haapasalo M, Lounatmaa K, Konttinen among periodontal pathogens, emphasizing Bacteroides
YT, Salo T, Grenier D, Sorsa T. Membrane components of gingivalis±Actinomyces viscosus cohesion on a saliva-
Treponema denticola trigger proteinase release from hu- coated mineral surface. Can J Microbiol 1988: 34: 299±306.
man polymorphonuclear leukocytes. J Dent Res 1996: 75: 64. Evans RT, Klausen B, Genco RJ. Immunization with fim-
1986±1993. brial protein and peptide protects against Porphyromonas
48. Du L, Pellen-Mussi P, Chandad F, Mouton C, Bonnaure- gingivalis-induced periodontal tissue destruction. Adv Exp
Mallet M. Conservation of fimbriae and the hemagglutinat- Med Biol 1992: 327: 255±262.
ing adhesin HA-Ag2 among Porphyromonas gingivalis 65. Evans RT, Klausen B, Ramamurthy NS, Golub LM, Sfintes-
strains and other anaerobic bacteria studied by epitope cu C, Genco RJ. Periodontopathic potential of two strains
mapping analysis. Clin Diagn Lab Immunol 1997: 4: of Porphyromonas gingivalis in gnotobiotic rats. Arch Oral
711±714. Biol 1992: 37: 813±819.

123
O'Brien-Simpson et al.

66. Fan Q, Sims TJ, Nakagawa T, Page RC. Antigenic cross- confers immunity against Porphyromonas gingivalis infec-
reactivity among Porphyromonas gingivalis serotypes. Oral tion in mice. Infect Immun 1998: 66: 4108±4114.
Microbiol Immunol 2000: 15: 158±165. 83. Genco RJ, Zambon JJ, Christersson LA. The origin of peri-
67. Fan Q, Sims T, Sojar H, Genco R, Page RC. Fimbriae of odontal infections. Adv Dent Res 1988: 2: 245±259.
Porphyromonas gingivalis induce opsonic antibodies that 84. Geysen HM, Meloen RH, Barteling SJ. Use of peptide
significantly enhance phagocytosis and killing by human synthesis to probe viral antigens for epitopes to a resolu-
polymorphonuclear leukocytes. Oral Microbiol Immunol tion of a single amino acid. Proc Natl Acad Sci USA 1984:
2001: 16: 144±152. 81: 3998±4002.
68. Farida R, Wilson M, Ivanyi L. Serum IgG antibodies to 85. Gibson FC, 3rd, Genco CA. Prevention of Porphyromonas
lipopolysaccharides in various forms of periodontal dis- gingivalis-induced oral bone loss following immunization
ease in man. Arch Oral Biol 1986: 31: 711±715. with gingipain R1. Infect Immun 2001: 69: 7959±7963.
69. Farquharson SI, Germaine GR, Gray GR. Isolation and 86. Gmu È r R, Hrodek K, Saxer UP, Guggenheim B. Double-blind
characterization of the cell-surface polysaccharides of Por- analysis of the relation between adult periodontitis and
phyromonas gingivalis ATCC 53978. Oral Microbiol Immu- systemic host response to suspected periodontal patho-
nol 2000: 15: 151±157. gens. Infect Immun 1986: 52: 768±776.
70. Fenno J, McBride B. Virulence factors of oral treponemes. 87. Gmur R, Strub JR, Guggenheim B. Prevalence of Bacter-
Anaerobe 1998: 4: 1±17. oides forsythus and Bacteroides gingivalis in subgingival
71. Fenno JC, Muller KH, McBride BC. Sequence analysis, ex- plaque of prosthodontically treated patients on short re-
pression, and binding activity of recombinant major outer call. J Periodontal Res 1989: 24: 113±120.
sheath protein (Msp) of Treponema denticola. J Bacteriol 88. Gmur R, McNabb H, van Steenbergen TJ, Baehni P,
1996: 178: 2489±2497. Mombelli A, van Winkelhoff AJ, Guggenheim B. Seroclas-
72. Fenno JC, Wong GW, Hannam PM, Muller KH, Leung WK, sification of hitherto nontypeable Actinobacillus actinomy-
McBride BC. Conservation of msp, the gene encoding the cetemcomitans strains: evidence for a new serotype e. Oral
major outer membrane protein of oral Treponema spp. Microbiol Immunol 1993: 8: 116±120.
J Bacteriol 1997: 179: 1082±1089. 89. Gopalsami C, Yotis W, Corrigan K, Schade S, Keene J,
73. Fenno JC, Hannam PM, Leung WK, Tamura M, Uitto VJ, Simonson L. Effect of outer membrane of Treponema den-
McBride BC. Cytopathic effects of the major surface pro- ticola on bone resorption. Oral Microbiol Immunol 1993: 8:
tein and the chymotrypsinlike protease of Treponema den- 121±124.
ticola. Infect Immun 1998: 66: 1869±1877. 90. Goulhen F, Hafezi A, Uitto VJ, Hinode D, Nakamura R,
74. Fenno JC, Lee SY, Bayer CH, Ning Y. The opdB locus en- Grenier D, Mayrand D. Subcellular localization and cyto-
codes the trypsin-like peptidase activity of Treponema den- toxic activity of the GroEL-like protein isolated from Actino-
ticola. Infect Immun 2001: 69: 6193±6200. bacillus actinomycetemcomitans. Infect Immun 1998: 66:
75. Firoozkoohi J, Zandi H, Olsen I. Comparison of lipopoly- 5307±5313.
saccharides from Bacteroides, Porphyromonas, Prevotella, 91. de Graeff-Meeder ER, Rijkers GT, Voorhorst-Ogink MM,
Campylobacter and Wolinella spp. by tricine±SDS±PAGE. Kuis W, van der Zee R, van Eden W, Zegers BJ. Antibodies
Endod Dent Traumatol 1997: 13: 13±18. to human HSP60 in patients with juvenile chronic arthritis,
76. Fishburn CS, Slaney JM, Carman RJ, Curtis MA. Degrada- diabetes mellitus, and cystic fibrosis. Pediatr Res 1993: 34:
tion of plasma proteins by the trypsin-like enzyme of Por- 424±428.
phyromonas gingivalis and inhibition of protease activity 92. Gregory RL, Kim DE, Kindle JC, Hobbs LC, Lloyd DR. Im-
by a serine protease inhibitor of human plasma. Oral Mi- munoglobulin-degrading enzymes in localized juvenile
crobiol Immunol 1991: 6: 209±251. periodontitis. J Periodontal Res 1992: 27: 176±183.
77. Fives-Taylor P, Meyer D, Mintz K. Characteristics of Acti- 93. Grenier D. Characteristics of hemolytic and hemaggluti-
nobacillus actinomycetemcomitans invasion of and adhe- nating activities of Treponema denticola. Oral Microbiol
sion to cultured epithelial cells. Adv Dent Res 1995: 9: 55±62. Immunol 1991: 6: 246±249.
78. Fives-Taylor PM, Meyer DH, Mintz KP, Brissette C. Viru- 94. Grenier D. Demonstration of a bimodal coaggregation re-
lence factors of Actinobacillus actinomycetemcomitans. action between Porphyromonas gingivalis and Treponema
Periodontol 2000 1999: 20: 136±167. denticola. Oral Microbiol Immunol 1992: 7: 280±284.
79. Fujiwara T, Ogawa T, Sobue S, Hamada S. Chemical, im- 95. Grenier D. Characterisation of the trypsin-like activity of
munobiological and antigenic characterizations of lipopo- Bacteroides forsythus. Microbiology 1995: 141: 921±926.
lysaccharides from Bacteroides gingivalis strains. J Gen 96. Grenier D. Degradation of host protease inhibitors and
Microbiol 1990: 136: 319±326. activation of plasminogen by proteolytic enzymes from
80. Gascan H, Gauchat JF, Aversa G, Van Vlasselaer P, de Vries Porphyromonas gingivalis and Treponema denticola. Mi-
JE. Anti-CD40 monoclonal antibodies or CD4‡ T cell crobiology 1996: 142: 955±961.
clones and IL-4 induce IgG4 and IgE switching in purified 97. Grenier D, Uitto VJ, McBride BC. Cellular location of a
human B cells via different signaling pathways. J Immunol Treponema denticola chymotrypsinlike protease and im-
1991: 147: 8±13. portance of the protease in migration through the base-
81. Gazi MI, Cox SW, Clark DT, Eley BM. Characterization of ment membrane. Infect Immun 1990: 58: 347±351.
protease activities in Capnocytophaga spp., Porphyromo- 98. Grenier D, Plamondon P, Sorsa T, Lee HM, McNamara T,
nas gingivalis, Prevotella spp., Treponema denticola and Ramamurthy NS, Golub LM, Teronen O, Mayrand D. In-
Actinobacillus actinomycetemcomitans. Oral Microbiol Im- hibition of proteolytic, serpinolytic, and progelatinase-b
munol 1997: 12: 240±248. activation activities of periodontopathogens by doxycy-
82. Genco CA, Odusanya BM, Potempa J, Mikolajczyk Pawlins- cline and the non-antimicrobial chemically modified tetra-
ka J, Travis J. A peptide domain on gingipain R which cycline derivatives. J Periodontol 2002: 73: 79±85.

124
 Antigens of bacteria associated with periodontitis

99. Gu K, Bainbridge B, Darveau RP, Page RC. Antigenic com- 114. Hirano H, Komatsu S, Kajiwara H, Takagi Y, Tsunasawa S.
ponents of Actinobacillus actinomycetemcomitans lipo- Microsequence analysis of the N-terminally blocked pro-
polysaccharide recognized by sera from patients with teins immobilized on polyvinylidene difluoride membrane
localized juvenile periodontitis. Oral Microbiol Immunol by western blotting. Electrophoresis 1993: 14: 839±846.
1998: 13: 150±157. 115. Hiroi M, Shimojima T, Kashimata M, Miyata T, Takano H,
100. Gunsolley JC, Tew JG, Gooss CM, Burmeister JA, Schenkein Takahama M, Sakagami H. Inhibition by Porphyromonas
HA. Effects of race and periodontal status on antibody gingivalis lipopolysaccharide of apoptosis induction in
reactive with Actinobacillus actinomycetemcomitans strain human peripheral blood polymorphonuclear leukocytes.
Y4. J Periodontal Res 1988: 23: 303±307. Anticancer Res 1998: 18: 3475±3479.
101. Haapasalo M, Muller KH, Uitto VJ, Leung WK, McBride BC. 116. Hirschfeld M, Weis JJ, Toshchakov V, Salkowski CA, Cody
Characterization, cloning, and binding properties of the MJ, Ward DC, Qureshi N, Michalek SM, Vogel SN. Signaling
major 53-kilodalton Treponema denticola surface antigen. by toll-like receptor 2 and 4 agonists results in differential
Infect Immun 1992: 60: 2058±2065. gene expression in murine macrophages. Infect Immun
102. Haapasalo M, Hannam P, McBride BC, Uitto VJ. Hyalur- 2001: 69: 1477±1482.
onan, a possible ligand mediating Treponema denticola 117. Holt SC, Bramanti TE. Factors in virulence expression and
binding to periodontal tissue. Oral Microbiol Immunol their role in periodontal disease pathogenesis. Crit Rev
1996: 11: 156±160. Oral Biol Med 1991: 2: 177±281.
103. Haffajee AD, Socransky SS. Microbial etiological agents of 118. Holt SC, Ebersole J, Felton M, Brunsvold M, Korman KS.
destructive periodontal diseases. Periodontol 2000 1994: 5: Implantation of Bacteroides gingivalis in non-human pri-
78±111. mates initiates progression of periodontitis. Science 1988:
104. Hajishengallis G, Martin M, Sojar HT, Sharma A, Schifferle 239: 55±57.
RE, DeNardin E, Russell MW, Genco RJ. Dependence of 119. Hongyo H, Kurihara H, Kokeguchi S, Miyamoto M, Maeda
bacterial protein adhesins on toll-like receptors for proin- H, Hayakawa M, Abiko Y, Takashiba S, Murayama Y.
flammatory cytokine induction. Clin Diagn Lab Immunol Molecular cloning and characterization of the gene encod-
2002: 9: 403±411. ing 53 kDa outer membrane protein of Porphyromonas
105. Hamada N, Watanabe K, Sasakawa C, Yoshikawa M, Yoshi- gingivalis. Microbios 1997: 92: 47±57.
mura F, Umemoto T. Construction and characterization of 120. Honma K, Kuramitsu HK, Genco RJ, Sharma A. Develop-
a fimA mutant of Porphyromonas gingivalis. Infect Immun ment of a gene inactivation system for Bacteroides for-
1994: 62: 1696±1704. sythus: construction and characterization of a BspA
106. Hanazawa S, Hirose K, Ohmori Y, Amano S, Kitano S. mutant. Infect Immun 2001: 69: 4686±4690.
Bacteroides gingivalis fimbriae stimulate production of 121. Hopp TP, Woods KR. Prediction of protein antigenic de-
thymocyte-activating factor by human gingival fibroblasts. terminants from amino acid sequences. Proc Natl Acad Sci
Infect Immun 1988: 56: 272±274. USA 1981: 787: 3824±3828.
107. Hanazawa S, Murakami Y, Hirose K, Amano S, Ohmori Y, 122. Humphery-Smith I, Cordwell SJ, Blackstock WP. Proteome
Higuchi H, Kitano S. Bacteroides (Porphyromonas) gingi- research: complementarity and limitations with respect
valis fimbriae activate mouse peritoneal macrophages and to the RNA and DNA worlds. Electrophoresis 1997: 18:
induce gene expression and production of interleukin-1. 1217±1242.
Infect Immun 1991: 59: 1972±1977. 123. Iino Y, Hopps RM. The bone-resorbing activities in tissue
108. Hanazawa S, Murakami Y, Takeshita A, Kitami H, Ohta K, culture of lipopolysaccharides from the bacteria Actinoba-
Amano S, Kitano S. Porphyromonas gingivalis fimbriae in- cillus actinomycetemcomitans, Bacteroides gingivalis and
duce expression of the neutrophil chemotactic factor KC Capnocytophaga ochracea isolated from human mouths.
gene of mouse peritoneal macrophages: role of protein Arch Oral Biol 1984: 29: 59±63.
kinase C. Infect Immun 1992: 60: 1544±1549. 124. Inoue T, Tanimoto I, Ohta H, Kato K, Murayama Y, Fukui
109. Hanley SA, Aduse-Opoku J, Curtis MA. A 55-kilodalton K. Molecular characterization of low-molecular-weight
immunodominant antigen of Porphyromonas gingivalis component protein, Flp, in Actinobacillus actinomycetem-
W50 has arisen via horizontal gene transfer. Infect Immun comitans fimbriae. Microbiol Immunol 1998: 42: 253±258.
1999: 67: 1157±1171. 125. Inoue T, Ohta H, Tanimoto I, Shingaki R, Fukui K. Hetero-
110. Harano K, Yamanaka A, Okuda K. An antiserum to a syn- geneous post-translational modification of Actinobacillus
thetic fimbrial peptide of Actinobacillus actinomycetemco- actinomycetemcomitans fimbrillin. Microbiol Immunol
mitans blocked adhesion of the microorganism. FEMS 2000: 44: 715±718.
Microbiol Lett 1995: 130: 279±285. 126. Ishihara Y, Nishihara T, Maki E, Noguchi T, Koga T. Role of
111. Hendtlass A, Dashper SG, Reynolds EC. Identification of an interleukin-1 and prostaglandin in in vitro bone resorption
antigenic protein Pga30 from Porphyromonas gingivalis induced by Actinobacillus actinomycetemcomitans lipopo-
W50. Oral Microbiol Immunol 2000: 15: 383±387. lysaccharide. J Periodontal Res 1991: 26: 155±160.
112. Henzel WJ, Billeci TM, Stults JT, Wong SC, Grimley C, 127. Ishihara K, Miura T, Kuramitsu HK, Okuda K. Character-
Watanabe C. Identifying proteins from two-dimensional ization of the Treponema denticola prtP gene encoding a
gels by molecular mass searching of peptide fragments prolyl-phenylalanine-specific protease (dentilisin). Infect
in protein sequence databases. Proc Natl Acad Sci USA Immun 1996: 64: 5178±5186.
1993: 90: 5011±5015. 128. Ishihara K, Kuramitsu HK, Miura T, Okuda K. Dentilisin
113. Hinode D, Nakamura R, Grenier D, Mayrand D. Cross- activity affects the organization of the outer sheath of Tre-
reactivity of specific antibodies directed to heat shock pro- ponema denticola. J Bacteriol 1998: 180: 3837±3844.
teins from periodontopathogenic bacteria and of human 129. Isogai H, Isogai E, Yoshimura F, Suzuki T, Kagota W,
origin (corrected). Oral Microbiol Immunol 1998: 13: 55±58. Takano K. Specific inhibition of adherence of an oral strain

125
O'Brien-Simpson et al.

of Bacteroides gingivalis 381 to epithelial cells by mono- 144. Katz J, Ward DC, Michalek SM. Effect of host response on
clonal antibodies against the bacterial fimbriae. Arch Oral the pathogenicity of strains of Porphyromonas gingivalis.
Biol 1988: 33: 479±485. Oral Microbiol Immunol 1996: 5: 309±318.
130. Isogai E, Isogal H, Kimura K, Fujii N, Takagi S, Hirose K, 145. Katz J, Black KP, Michalek SM. Host responses to
Hayashi M. In vivo induction of apoptosis and immune recombinant hemagglutinin B of Porphyromonas gingi-
responses in mice by administration of lipopolysaccharide valls in an experimental rat model. Infect Immun 1999:
from Porphyromonas gingivalis. Infect Immun 1996: 64: 67: 4352±4359.
1461±1466. 146. Kawai T, Eisen-Lev R, Seki M, Eastcott JW, Wilson
131. Ito H, Shuto T, Takada H, Koga T, Aida Y, Hirata M. Lipo- ME, Taubman MA. Requirement of B7 costimulation
polysaccharides from Porphyromonas gingivalis, Prevotella for Th1-mediated inflammatory bone resorption in
intermedia and Actinobacillus actinomycetemcomitans experimental periodontal disease. J Immunol 2000: 164:
promote osteoclastic differentiation in vitro. Arch Oral Biol 2102±2109.
1996: 41: 439±444. 147. Kawata Y, Hanazawa S, Amano S, Murakami Y, Matsumoto
132. Iwase M, Korchak HM, Lally ET, Berthold P, Taichman NS. T, Nishida K, Kitano S. Porphyromonas gingivalis fimbriae
Lytic effects of Actinobacillus actinomycetemcomitans stimulate bone resorption in vitro. Infect Immun 1994: 62:
leukotoxin on human neutrophil cytoplasts. J Leukoc Biol 3012±3016.
1992: 52: 224±227. 148. Kelly CG, Booth H, Kendal JM, Slaney MA, Curtis MA,
133. Jagels MA, Travis J, Potempa J, Pike R, Hugh TE. Proteolytic Lehner T. The relationship between colonisation and hea-
inactivation of the leukocyte C5a receptor by proteinases magglutination inhibiting and B cell epitopes of P. gingi-
derived from Porphyromonas gingivalis. Infect Immun valis. Clin Exp Immunol 1997: 110: 285±291.
1996: 64: 1984±1991. 149. Kent LW, Rahemtulla F, Michalek SM. Interleukin (IL)-1
134. Jammerson BA, Wolf H. Predicting antigenicity from pro- and Porphyromonas gingivalis lipopolysaccharide stimula-
tein primary structure: a new algorthm for the prediction of tion of IL-6 production by fibroblasts derived from healthy
antigenic sites. CABIOS 1988: 4: 181. or periodontally diseased human gingival tissue. J Period-
135. Jonas D, Schultheis B, Klas C, Krammer PH, Bhakdi S. ontol 1999: 70: 274±282.
Cytocidal effects of Escherichia coli hemolysin on human 150. Kesavalu L, Holt SC, Ebersole JL. Porphyromonas gingivalis
T lymphocytes. Infect Immun 1993: 61: 1715±1721. virulence in a murine lesion model: effects of immune
136. Jonas D, Walev I, Berger T, Liebetrau M, Palmer M, Bhakdi alterations. Microb Pathog 1997: 23: 317±326.
S. Novel path to apoptosis: small transmembrane pores 151. Kesavalu L, Walker SG, Holt SC, Crawley RR, Ebersole JL.
created by staphylococcal alpha-toxin in T lymphocytes Virulence characteristics of oral treponemes in a murine
evoke internucleosomal DNA degradation. Infect Immun model. Infect Immun 1997: 65: 5096±5102.
1994: 62: 1304±1312. 152. Kesavalu L, Holt S, Ebersole J. Virulence of a polymicrobic
137. Jotwani R, Palucka AK, Al-Quotub M, Nouri-Shirazi M, Kim complex, Treponema denticola and Porphyromonas gingi-
J, Bell D, Banchereau J, Cutler CW. Mature dendritic cells valis, in a murine model. Oral Microbiol Immunol 1998: 13:
infiltrate the T cell-rich region of oral mucosa in chronic 373±377.
periodontitis: in situ, in vivo, and in vitro studies. J Immu- 153. Kesavalu L, Holt SC, Ebersole JL. Lack of humoral immune
nol 2001: 167: 4693±4700. protection against Treponema denticola virulence in a
138. Jungblut PR, Grabher G, Stoffler G. Comprehensive detec- murine model. Infect Immun 1999: 67: 5736±5746.
tion of immunorelevant Borrelia garinii antigens by two- 154. Keulers RA, Maltha JC, Mikx FH, Wolters-Lutgerhorst JM.
dimensional electrophoresis. Electrophoresis 1999: 20: Involvement of treponemal surface-located protein and
3611±3622. carbohydrate moieties in the attachment of Treponema
139. Kadowaki T, Yoneda M, Okamoto K, Maeda K, Yamamoto denticola ATCC 33520 to cultured rat palatal epithelial
K. Purification and characterization of a novel arginine- cells. Oral Microbiol Immunol 1993: 8: 236±241.
specific cysteine proteinase (argingipain) involved in the 155. Khlgatian M, Nassar H, Chou HH, Gibson FC, 3rd, Genco
pathogenesis of periodontal disease from the culture CA. Fimbria-dependent activation of cell adhesion mole-
supernatant of Porphyromonas gingivalis. J Biol Chem cule expression in Porphyromonas gingival-infected en-
1994: 269: 21371±21378. dothelial cells. Infect Immun 2002: 70: 257±267.
140. Kaplan JB, Perry MB, MacLean LL, Furgang D, Wilson ME, 156. Kimsey RB, Spielman A. Motility of Lyme disease spiro-
Fine DH. Structural and genetic analyses of O polysacchar- chetes in fluids as viscous as the extracellular matrix.
ide from Actinobacillus actinomycetemcomitans serotype f. J Infect Dis 1990: 162: 1205±1208.
Infect Immun 2001: 69: 5375±5384. 157. Kinder SA, Holt SC. Characterization of coaggregation be-
141. Kataoka M, Kawamura K, Kondoh T, Wakano Y, Ishida H. tween Bacteroides gingivalis T22 and Fusobacterium nucle-
Purification of a fibroblast-inhibitory factor from Actino- aturn T18. Infect Immun 1989: 57: 3425±3433.
bacillus actinomycetemcomitans Y4. FEMS Microbiol Lett 158. Kinder SA, Holt SC. Coaggregation between bacterial spe-
1993: 107: 111±114. cies. Methods Enzymol 1994: 236: 254±270.
142. Kato T, Okuda K. Actinobacillus actinomycetemcomitans 159. Kirby AC, Meghji S, Nair SP, White P, Reddi K, Nishihara T,
possesses an antigen binding to anti-human IL-10 anti- Nakashima K, Willis AC, Sim R, Wilson M, et al. The potent
body. FEMS Microbiol Lett 2001: 204: 293±297. bone-resorbing mediator of Actinobacillus actinomycetem-
143. Kato T, Honma K, Yamanaka A, Miura T, Okuda K. Hetero- comitans is homologous to the molecular chaperone
geneity in the immune response to serotype b lipopolysac- GroEL. J Clin Invest 1995: 96: 1185±1194.
charide of Actinobacillus actinomycetemcomitans in inbred 160. Kitani A, Strober W. Regulation of C gamma subclass
strains of mice. FEMS Immunol Med Microbiol 2000: 28: germ-line transcripts in human peripheral blood B cells.
67±70. J Immunol 1993: 151: 3478±3488.

126
 Antigens of bacteria associated with periodontitis

161. Klitorinos A, Noble P, Siboo R, Chan EC. Viscosity-depen- 177. Kwok W, Gebe J, Liu A, Agar S, Ptacek N, Hammer J, Koelle
dent locomotion of oral spirochetes. Oral Microbiol Im- D, Nepom G. Rapid epitope identification from complex
munol 1993: 8: 242±244. class-II-restricted T-cell antigens. Trends Immunol 2001:
162. Kobe B, Deisenhofer J. The leucine-rich repeat: a versatile 22: 583±588.
binding motif. Trends Biochem Sci 1994: 19: 415±421. 178. Kyte J, Doolittle RF. A simple method for displaying the
163. Koga T, Nishihara T, Fujiwara T, Nisizawa T, Okahashi N, hydropathic character of a protein. J Mol Biol 1982: 157:
Noguchi T, Hamada S. Biochemical and immunobiological 105±132.
properties of lipopolysaccharide (LPS) from Bacteroides 179. Laine ML, Appelmelk BJ, van Winkelhoff AJ. Novel poly-
gingivalis and comparison with lipopolysaccharide from saccharide capsular serotypes in Porphyromonas gingiva-
Escherichia coli. Infect Immun 1985: 47: 638±647. lis. J Periodontal Res 1996: 31: 278±284.
164. Koga T, Kusuzaki T, Asakawa H, Senpuku H, Nishihara T, 180. Laine ML, van Winkelhoff AJ. Virulence of six capsular
Noguchi T. The 64-kilodalton GroEL-like protein of Acti- serotypes of Porphyromonas gingivalis in a mouse model.
nobacillus actinomycetemcomitans. J Periodontal Res 1993: Oral Microbiol Immunol 1998: 13: 322±325.
28: 475±477. 181. Lally ET, Golub EE, Kieba IR, Taichman NS, Rosenbloom J,
165. Kohler JJ, Pathangey LB, Brown TA. Oral immunization with Rosenbloom JC, Gibson CW, Demuth DR. Analysis of the
recombinant Salmonella typhimurium expressing a cloned Actinobacillus actinomycetemcomitans leukotoxin gene.
Porphyromonas gingivalis hemagglutinin: effect of boosting Delineation of unique features and comparison to homo-
on mucosal, systemic and immunoglobulin G subclass logous toxins. J Biol Chem 1989: 264: 15451±15456.
response. Oral Microbiol Immunol 1998: 13: 81±88. 182. Lally ET, Kieba IR, Demuth DR, Rosenbloom J, Golub EE,
166. Kokeguchi S, Miyamoto M, Kato K, Tanimoto I, Kurihara H, Taichman NS, Gibson CW. Identification and expression of
Murayama Y. Isolation and characterization of a 53 kDa the Actinobacillus actinomycetemcomitans leukotoxin
major cell envelope protein antigen from Treponema den- gene. Biochem Biophys Res Commun 1989: 159: 256±262.
ticola ATCC 35405. J Periodontal Res 1994: 29: 70±78. 183. Lally ET, Golub EE, Kieba IR. Identification and immuno-
167. Kolenbrander PE, Andersen RN. Inhibition of coaggrega- logical characterization of the domain of Actinobacillus
tion between Fusobacterium nucleatum and Porphyromo- actinomycetemcomitans leukotoxin that determines its
nas (Bacteroides) gingivalis by lactose and related sugars. specificity for human target cells. J Biol Chem 1994: 269:
Infect Immun 1989: 57: 3204±3209. 31289±31295.
168. Kolenbrander PE, Andersen RN, Holdeman LV. Coaggrega- 184. Lally ET, Kieba IR, Sato A, Green CL, Rosenbloom J,
tion of oral Bacteroides species with other bacteria: central Korostoff J, Wang JF, Shenker BJ, Ortlepp S, Robinson
role in coaggregation bridges and competitions. Infect Im- MK, Billings PC. RTX toxins recognize a beta2 integrin on
mun 1985: 48: 741±746. the surface of human target cells. J Biol Chem 1997: 272:
169. Kolenbrander PE, Parrish KD, Andersen RN, Greenberg EP. 30463±30469.
Intergeneric coaggregation of oral Treponema spp. with 185. Lamont RJ, Chan A, Belton CM, Izutsu KT, Vasel D, Wein-
Fusobacterium spp. and intrageneric coaggregation among berg A. Porphyromonas gingivalis invasion of gingival
Fusobacteriurn spp. Infect Immun 1995: 63: 4584±4588. epithelial cells. Infect Immun 1995: 63: 3878±3885.
170. Korostoff J, Wang JF, Kieba I, Miller M, Shenker BJ, Lally 186. Lantz MS, Rowland RW, Switalski LM, Hook M. Interac-
ET. Actinobacillus actinomycetemcomitans leukotoxin in- tions of Bacteroides gingivalis with fibrinogen. Infect Im-
duces apoptosis in HL-60 cells. Infect Immun 1998: 66: mun 1986: 54: 654±658.
4474±4483. 187. Laosrisin N, Nakashima K, Ishikawa I. Detection of Bacter-
171. Krupka HI, Huber R, Holt SC, Clausen T. Crystal structure oides gingivalis antigenic proteins by immunoblotting ana-
of cystalysin from Treponema denticola: a pyridoxal 5'- lysis. J Periodontol 1990: 61: 261±268.
phosphate-dependent protein acting as a haemolytic en- 188. Larjava H, Uitto VJ, Haapasalo M, Heino J, Vuento M.
zyme. Embo J 2000: 19: 3168±3178. Fibronectin fragmentation induced by dental plaque and
172. Kumada H, Kondo S, Umemoto T, Hisatsune K. Chemical Bacteroides gingivalis. Scand J Dent Res 1987: 95: 308±314.
structure of the 2-keto-3-deoxyoctonate region of lipo- 189. Lee JY, Sojar HT, Bedi GS, Genco RJ. Porphyromonas (Bac-
polysaccharide isolated from Porphyromonas (Bacteroides) teroides) gingivalis fimbrillin: size, amino-terminal se-
gingivalis. FEMS Microbiol Lett 1993: 108: 75±79. quence, and antigenic heterogeneity. Infect Immun 1991:
173. Kumada H, Haishima Y, Umemoto T, Tanamoto K. Struc- 59: 383±389.
tural study on the free lipid A isolated from lipopolysac- 190. Lee JY, Sojar HT, Bedi GS, Genco RJ. Synthetic peptides
charide of Porphyromonas gingivalis. J Bacteriol 1995: 177: analogous to the fimbrillin sequence inhibit adherence
2098±2106. of Porphyromonas gingivalis. Infect Immun 1992: 60:
174. Kurihara H, Nishimura F, Nakamura T, Nakagawa M, 1662±1670.
Tanimoto I, Nomura Y, Kokeguchi S, Kato K, Murayama 191. Lepine G, Ellen RP, Progulske-Fox A. Construction and
Y. Humoral immune response to an antigen from Porphyr- preliminary characterization of three hemagglutinin mu-
omonas gingivalis 381 in periodontal disease. Infect Im- tants of Porphyromonas gingivalis. Infect Immun 1996: 64:
mun 1991: 59: 2758±2762. 1467±1472.
175. Kurita-Ochiai T, Ochiai K. Immunosuppressive factor from 192. Lepine G, Progulske-Fox A. Duplication and differential
Actinobacillus actinomycetemcomitans down regulates expression of hemagglutinin genes in Porphyromonas gin-
cytokine production. Infect Immun 1996: 64: 50±54. givalis. Oral Microbiol Immunol 1996: 11: 65±78.
176. Kurzban GP, Chu L, Ebersole JL, Holt SC. Sulfhemoglobin 193. Lind SM, Kenne L, Lindberg AA. Mapping of the binding
formation in human erythrocytes by cystalysin, an L-cys- specificity for five monoclonal antibodies recognizing 3-
teine desulfhydrase from Treponema denticola. Oral Mi- deoxy-D-manno-octulosonic acid in bacterial lipopolysac-
crobiol Immunol 1999: 14: 153±164. charides. J Immunol 1991: 146: 3864±3870.

127
O'Brien-Simpson et al.

194. Ling TY, Sims TJ, Chen HA, Whitney CW, Moncla BJ, Engel ATCC 35405: evidence of hydrolysis of human bioactive
LD, Page RC. Titer and subclass distribution of serum IgG peptides. Infect Immun 1994: 62: 4938±4947.
antibody reactive with Actinobacillus actinomycetemcomi- 211. Makinen KK, Makinen PL, Loesche WJ, Syed SA. Puri-
tans in localized juvenile periodontitis. J Clin Immunol fication and general properties of an oligopeptidase
1993: 13: 101±112. from Treponema denticola ATCC 35405 ± a human oral
195. Listgarten MA, Lai CH, Evian CI. Comparative antibody spirochete. Arch Biochem Biophys 1995: 316: 689±698.
titers to Actinobacillus actinomycetemcomitans in juvenile 212. Makinen PL, Makinen KK, Syed SA. Role of the chymotryp-
periodontitis, chronic periodontitis and periodontally sin-like membrane-associated proteinase from Treponema
healthy subjects. J Clin Periodontol 1981: 8: 155±164. denticola ATCC 35405 in inactivation of bioactive peptides.
196. Loesche WJ. Chemotherapy of dental plaque infections. Infect Immun 1995: 63: 3567±3575.
Oral Sci Rev 1976: 9: 65±107. 213. Makinen KK, Chen CY, Makinen PL. Proline iminopepti-
197. Loesche WJ. Possibilities for treating periodontal disease as dase from the outer cell envelope of the human oral spiro-
specific anaerobic infections. J Can Dent Assoc 1984: 50: chete Treponema denticola ATCC 35405. Infect Immun
467±472. 1996: 64: 702±708.
198. Loesche WJ, Syed SA, Stoll J. Trypsin-like activity in sub- 214. Mangan DF, Taichman NS, Lally ET, Wahl SM. Lethal
gingival plaque. A diagnostic marker for spirochetes and effects of Actinobacillus actinomycetemcomitans leuko-
periodontal disease? J Periodontol 1987: 58: 266±273. toxin on human T lymphocytes. Infect Immun 1991: 59:
199. Lopatin DE, Blackburn E. Avidity and titer of immunoglo- 3267±3272.
bulin G subclasses to Porphyromonas gingivalis in adult 215. Masuda K, Kawata T. Isolation, properties, and reassembly
periodontitis patients. Oral Microbiol Immunol 1992: 7: of outer sheath carrying a polygonal array from an oral
332±337. treponeme. J Bacteriol 1982: 150: 1405±1413.
200. Lourbakos A, Potempa J, Travis J, D'Andrea MR, Andrade- 216. Mathers DA, Leung WK, Fenno JC, Hong Y, McBride BC.
Gordon P, Santulli R, Mackie EJ, Pike RN. Arginine-specific The major surface protein complex of Treponema denticola
protease from Porphyromonas gingivalis activates pro- depolarizes and induces ion channels in HeLa cell mem-
tease-activated receptors on human oral epithelial cells branes. Infect Immun 1996: 64: 2904±2910.
and induces interleukin-6 secretion. Infect Immun 2001: 217. McKee AS, McDermid AS, Wait R, Baskerville A. Isolation of
69: 5121±5130. colonial variants of Bacteroides gingivalis W50 with re-
201. Lourbakos A, Yuan YP, Jenkins AL, Travis J, Andrade- duced virulence. J Med Microbiol 1988: 27: 59±64.
Gordon P, Santulli R, Potempa J, Pike RN. Activation of 218. Meghji S, Wilson M, Henderson B, Kinane D. Anti-prolif-
protease-activated receptors by gingipains from Porphyro- erative and cytotoxic activity of surface-associated material
monas gingivalis leads to platelet aggregation: a new trait from periodontopathogenic bacteria. Arch Oral Biol 1992:
in microbial pathogenicity. Blood 2001: 97: 3790±3797. 37: 637±644.
202. Lu J, Celis E. Use of two predictive algorithms of the world 219. Meghji S, Wilson M, Barber P, Henderson B. Bone resorb-
wide web for the identification of tumor-reactive T-cell ing activity of surface-associated material from Actinoba-
epitopes. Cancer Res 2000: 60: 5223±5227. cillus actinomycetemcomitans and Eikenella corrodens. J
203. Lu H, Califano JV, Schenkein HA, Tew JG. Immunoglobulin Med Microbiol 1994: 41: 197±203.
class and subclass distribution of antibodies reactive with 220. Messner P. Bacterial glycoproteins. Glycoconjugate J 1997:
the immunodominant antigen of Actinobacillus actinomy- 14: 3±11.
cetemcomitans serotype b. Infect Immun 1993: 61: 2400± 221. Mikx FH, Jacobs F, Satumalay C. Cell-bound pep-
2407. tidase activities of Treponema denticola ATCC 33520
204. Lundgren M, Persson U, Larsson P, Magnusson C, Smith in continuous culture. J Gen Microbiol 1992: 138: 1837±
CI, Hammarstrom L, Severinson E. Interleukin 4 induces 1842.
synthesis of IgE and IgG4 in human B cells. Eur J Immunol 222. Millar SJ, Goldstein EG, Levine MJ, Hausmann E. Modula-
1989: 19: 1311±1315. tion of bone metabolism by two chemically distinct lipo-
205. Macnab RM. Genetics and biogenesis of bacterial flagella. polysaccharide fractions from Bacteroides gingivalis. Infect
Annu Rev Genet 1992: 26: 131±158. Immun 1986: 51: 302±306.
206. Maeda H, Miyamoto M, Kokeguchi S, Kono T, Nishimura 223. Millar DJ, Scott EE, Slaney JMUS, Benjamin P, Curtis MA.
F, Takashiba S, Murayama Y. Epitope mapping of heat Production and characterisation of monoclonal antibodies
shock protein 60 (GroEL) from Porphyromonas gingivalis. to the principle sonicate antigens of Porphyromonas
FEMS Immunol Med Microbiol 2000: 28: 219±224. gingivalis W50. FEMS Immunol Med Microbiol 1993: 7:
207. Maeji N, Bray A, Geysen H. Multi-pin peptide synthesis 211±222.
strategy for T cell determinant analysis. J Immunol Meth- 224. Miyata Y, Takeda H, Kitano S, Hanazawa S. Porphyromo-
ods 1990: 134: 23±33. nas gingivalis lipopolysaccharide-stimulated bone resorp-
208. Makinen KK, Makinen PL. The peptidolytic capacity of the tion via CD14 is inhibited by broad-spectrum antibiotics.
spirochete system. Med Microbiol Immunol (Berl) 1996: Infect Immun 1997: 65: 3513±3519.
185: 1±10. 225. Modrow S, Wolf H. Use of synthetic peptides as diagnostic
209. Makinen KK, Chen CY, Makinen PL, Ohta K, Loesche WJ. reagents in virology. In: van Regenmortel, MHV, Neurath,
The benzoylarginine peptidase from Treponema denticola R, editors. Immunochemistry of Viruses II. Amsterdam:
(strain ASLM), a human oral spirochaete: evidence for Elsevier Science, 1990: 83±101.
active-site carboxyl groups. Mol Microbiol 1990: 4: 226. Molloy MP, Herbert BR, Slade MB, Rabilloud T, Nouwens
1413±1417. AS, Williams KL, Gooley AA. Proteomic analysis of the
210. Makinen PL, Makinen KK, Syed SA. An endo-acting pro- Escherichia coli outer membrane. Eur J Biochem 2000:
line-specific oligopeptidase from Treponema denticola 267: 2871±2881.

128
 Antigens of bacteria associated with periodontitis

227. Mouton C, Bouchard D, Deslauriers M, Lamonde L. Im- 241. Newman MG, Socransky SS, Savitt ED, Propas DA,
munochemical identification and preliminary characteri- Crawford A. Studies of the microbiology of periodontosis.
zation of a nonfimbrial hemagglutinating adhesin of J Periodontol 1976: 47: 373±379.
Bacteroides gingivalis. Infect Immun 1989: 57: 566±573. 242. Nilsson I, Utt M, Nilsson HO, Ljungh A, Wadstrom T. Two-
228. Mouton C, Ni Eidhin D, Deslauriers M, Lamy L. The he- dimensional electrophoretic and immunoblot analysis of
magglutinating adhesin HA-Ag2 of Bacteroides gingivalis cell surface proteins of spiral-shaped and coccoid forms of
is distinct from fimbrilin. Oral Microbiol Immunol 1991: Helicobacter pylori. Electrophoresis 2000: 21: 2670±2677.
6: 6±11. 243. Nishida E, Hara Y, Kaneko T, Ikeda Y, Ukai T, Kato I. Bone
229. Muller D, Poolman JT, Bernadina WE, van Kol PJ, Ruiten- resorption and local interleukin-1 alpha and interleukin-
berg EJ. Characterization of outer membrane proteins from 1beta synthesis induced by Actinobacillus actinomycetem-
Actinobacillus actinomycetemcomitans. Microb Pathog comitans and Porphyromonas gingivalis lipopolysacchar-
1990: 9: 227±233. ide. J Periodontal Res 2001: 36: 1±8.
230. Munemasa T, Takemoto T, Dahlen G, Hino T, Shiba H, 244. Nishihara T, Koga T, Hamada S. Suppression of murine
Ogawa T, Kurihara H. Adherence of Bacteroides forsythus to macrophage interleukin-1 release by the polysaccharide
host cells. Microbios 2000: 101: 115±126. portion of Haemophilus actinomycetemcomitans lipopoly-
231. Murakami Y, Hanazawa S, Nishida K, Iwasaka H, Kitano S. saccharide. Infect Immun 1988: 56: 619±625.
N-acetyl-D-galactosamine inhibits TNF-alpha gene ex- 245. Nishihara T, Ueda N, Amano K, Ishihara Y, Hayakawa H,
pression induced in mouse peritoneal macrophages by Kuroyanagi T, Ohsaki Y, Nagata K, Noguchi T. Actinobacil-
fimbriae of Porphyromonas (Bacteroides) gingivalis, an lus actinomycetemcomitans Y4 capsular-polysaccharide-
oral anaerobe. Biochem Biophys Res Commun 1993: 192: like polysaccharide promotes osteoclast-like cell formation
826±832. by interleukin-1 alpha production in mouse marrow cul-
232. Nagata T, Bellows GC, Kasugai S, Butler WT, Sodek J. tures. Infect Immun 1995: 63: 1893±1898.
Biosynthesis of bone proteins [SPP±1 (secreted phospho- 246. Njoroge T, Genco RJ, Sojar HT, Hamada N, Genco CA. A
protein±1, osteopontin), BSP (bone sialoprotein) and role for fimbriae in Porphyromonas gingivalis invasion of
SPARC (osteonectin)] in association with mineralised-tis- oral epithelial cells. Infect Immun 1997: 65: 1980±1984.
sue formation by fetal-rat calvarial cells in culture. Bio- 247. Nouwens AS, Cordwell SJ, Larsen MR, Molloy MP, Gillings
chemistry 1991: 274: 513±520. M, Willcox MD, Walsh BJ. Complementing genomics with
233. Nair BC, Mayberry WR, Dziak R, Chen PB, Levine MJ, proteomics: the membrane subproteome of Pseudomonas
Hausmann E. Biological effects of a purified lipopolysac- aeruginosa PAO1. Electrophoresis 2000: 21: 3797±3809.
charide from Bacteroides gingivalis. J Periodontal Res 1983: 248. Novotny J, Handschumacher M, Haber E, Bruccoleri RE,
18: 40±49. Carlson WB, Fanning DW, Smith JA, Rose GD. Antigenic
234. Nakagawa I, Amano A, Kimura RK, Nakamura T, Kawabata determinants in proteins coincide with surface regions
S, Hamada S. Distribution and molecular characterization accessible to large probes (antibody domains). Proc Natl
of Porphyromonas gingivalis carrying a new type of fimA Acad Sci USA 1986: 83: 226±230.
gene. J Clin Microbiol 2000: 38: 1909±1914. 249. Obeid P, Bercy P. Effects of smoking on periodontal health:
235. Nakagawa T, Sims T, Fan Q, Potempa J, Travis J, Houston a review. Adv Ther 2000: 17: 230±237.
L, Page RC. Functional characteristics of antibodies in- 250. O'Brien-Simpson N, Paolini R, Reynolds E. RgpA-Kgp pep-
duced by Arg-gingipain (HRgpA) and Lys-gingipain (Kgp) tide-based immunogens provide protection against Por-
from Porphyromonas gingivalis. Oral Microbiol Immunol phyromonas gingivalis challenge in a murine lesion
2001: 16: 202±211. model. Infect Immun 2000: 68: 4055±4063.
236. Nakagawa I, Amano A, Kuboniwa M, Nakamura T, Kawa- 251. O'Brien-Simpson NM, Black CL, Bhogal PS, Cleal SM,
bata S, Hamada S. Functional differences among FimA Slakeski N, Higgins TJ, Reynolds EC. Serum IgG and IgG
variants of Porphyromonas gingivalis and their effects on subclass responses to the RgpA-Kgp proteianse±adhesin
adhesion to and invasion of human epithelial cells. Infect complex of P. gingivalis in adult periodontitis. Infect Im-
Immun 2002: 70: 277±285. mun 2000: 68: 2704±2712.
237. Nakamura Y, Umemoto T, Nakatani Y, Namikawa I, Wadood 252. O'Brien-Simpson NM, Paolini RA, Hoffmann B, Slakeski N,
A. Common and specific antigens of several treponemes Dashper SG, Reynolds EC. Role of RgpA, RgpB, and Kgp
detected by polyclonal antisera against major cellular proteinases in virulence of Porphyromonas gingivalis
proteins. Oral Microbiol Immunol 1993: 8: 288±294. W50 in a murine lesion model. Infect Immun 2001: 69:
238. Nakano Y, Inai Y, Yamashita Y, Nagaoka S, Kusuzaki- 7527±7534.
Nagira T, Nishihara T, Okahashi N, Koga T. Molecular 253. Ogawa T. Chemical structure of lipid A from Porphyromo-
and immunological characterization of a 64-kDa protein nas (Bacteroides) gingivalis lipopolysaccharide. FEBS Lett
of Actinobacillus actinomycetemcomitans. Oral Microbiol 1993: 332: 197±201.
Immunol 1995: 10: 151±159. 254. Ogawa T. The potential protective immune responses to
239. Nakayama K, Kadowaki T, Okamoto K, Yamamoto K. Con- synthetic peptides containing conserved epitopes of Por-
struction and characterization of arginine-specific cysteine phyromonas gingivalis fimbrial protein. J Med Microbiol
proteinase (Arg-gingipain)-deficient mutants of Porphyro- 1994: 41: 349±358.
monas gingivalis. Evidence for significant contribution of 255. Ogawa T, Shimauchi H, Hamada S. Mucosal and systemic
Arg-gingipain to virulence. J Biol Chem 1995: 270: 23619± immune responses in BALB/c mice to Bacteroides gingiva-
23626. lis fimbriae administered orally. Infect Immun 1989: 57:
240. Neiders ME, Chen PB, Suido H, Reynolds HS, Zambon JJ, 3466±3471.
Shlossman M, Genco RJ. Heterogeneity of virulence of 256. Ogawa T, Kusumoto Y, Hamada S, McGhee JR, Kiyono H.
Bacteroides gingivalis. J Periodontol Res 1989: 24: 192±198. Bacteroides gingivalis-specific serum IgG and IgA subclass

129
O'Brien-Simpson et al.

antibodies in periodontal diseases. Clin Exp Immunol 271. Persson GR, Engel D, Whitney G, Darveau R, Weinberg A,
1990: 82: 318±325. Brunsvold M, Page RC. Immunization against Porphyro-
257. Ogawa T, Kusumoto Y, Uchida H, Nagashima S, Ogo H, monas gingivalis inhibits progression of experimental per-
Hamada S. Immunobiological activities of synthetic pep- iodontitis in non-human primates. Infect Immun 1994: 62:
tide segments of fimbrial protein from Porphyromonas 1026±1031.
gingivalis. Biochem Biophys Res Commun 1991: 180: 272. Pfeilschifter J, Chenu C, Bird A, Mundy GR, Roodman GD.
1335±1341. Interleukin-1 and tumor necrosis factor stimulate the for-
258. Ogawa T, Uchida H, Hamada S. Porphyromonas gingivalis mation of human osteoclastlike cells in vitro. J Bone Miner
fimbriae and their synthetic peptides induce proinflamma- Res 1989: 4: 113±118.
tory cytokines in human peripheral blood monocyte cul- 273. Phadke ND, Molloy MP, Steinhoff SA, Ulintz PJ, Andrews
tures. FEMS Microbiol Lett 1994: 116: 237±242. PC, Maddock JR. Analysis of the outer membrane pro-
259. Ogawa T, Yasuda K, Yamada K, Mori H, Ochiai K, teome of Caulobacter crescentus by two-dimensional elec-
Hasegawa M. Immunochemical characterisation and epi- trophoresis and mass spectrometry. Proteomics 2001: 1:
tope mapping of a novel fimbrial protein (Pg-II fimbria) of 705±720.
Porphyromonas gingivalis. FEMS Immunol Med Microbiol 274. Pietrantonio F, Noble PB, Amsel R, Chan EC. Locomotory
1995: 11: 247±255. characteristics of Treponema denticola. Can J Microbiol
260. Ohta K, Makinen KK, Loesche WJ. Purification and char- 1988: 34: 748±752.
acterization of an enzyme produced by Treponema denti- 275. Pike RN, Potempa J, McGraw W, Coetzer THT, Travis J.
cola capable of hydrolyzing synthetic trypsin substrates. Characterization of the binding activities of proteinase±
Infect Immun 1986: 53: 213±220. adhesin complexes from Porphyromonas gingivalis. J Bac-
261. Ohyama H, Matsushita S, Kato N, Nishimura F, Oyaizu K, teriol 1996: 178: 2876±2882.
Kokeguchi S, Kurihara H, Takashiba S, Nishimura Y, 276. Potempa J, Banbula A, Travis J. Role of bacterial protei-
Murayama Y. T cell responses to 53-kDa outer membrane nases in matrix destruction and modulation of host res-
protein of Porphyromonas gingivalis in humans with early- ponses. Periodontol 2000 2000: 24: 153±192.
onset periodontitis. Hum Immunol 1998: 59: 635±643. 277. Preshaw PM, Schifferle RE, Walters JD. Porphyromonas
262. Onagawa M, Ishihara K, Okuda K. Coaggregation between gingivalis lipopolysaccharide delays human polymorpho-
Porphyromonas gingivalis and Treponema denticola. Bull nuclear leukocyte apoptosis in vitro. J Periodontal Res
Tokyo Dent Coll 1994: 35: 171±181. 1999: 34: 197±202.
263. Oyaizu K, Ohyama H, Nishimura F, Kurihara H, Matsushita 278. Progulske-Fox A, Tumwasorn S, Holt SC. The expression
S, Maeda H, Kokeguchi S, Hongyo H, Takashiba S, Mura- and function of a Bacteroides gingivalis hemagglutinin
yama Y. Identification and characterization of B-cell epi- gene in Escherichia coli. Oral Microbiol Immunol 1989: 4:
topes of a 53-kDa outer membrane protein from 121±131.
Porphyromonas gingivalis. Oral Microbiol Immunol 2001: 279. Progulske-Fox A, Tumwasorn S, Lepine G, Whitlock J,
16: 73±78. Savett D, Ferretti JJ, Banas JA. The cloning, expression
264. Page RC. Milestones in periodontal research and the re- and sequence analysis of a second Porphyromonas
maining critical issues. J Periodontal Res 1999: 34: 331±339. gingivalis gene that codes for a protein involved in
265. Paju S, Goulhen F, Asikainen S, Grenier D, Mayrand D, hemagglutination. Oral Microbiol Immunol 1995: 10:
Uitto V. Localization of heat shock proteins in clinical 311±318.
Actinobacillus actinomycetemcomitans strains and their ef- 280. Pulendran B, Kumar P, Cutler CW, Mohamadzadeh M, Van
fects on epithelial cell proliferation. FEMS Microbiol Lett Dyke T, Banchereau J. Lipopolysaccharides from distinct
2000: 182: 231±235. pathogens induce different classes of immune responses
266. Paramonov N, Bailey D, Rangarajan M, Hashim A, Kelly in vivo. J Immunol 2001: 167: 5067±5076.
G, Curtis MA, Hounsell EF. Structural analysis of the 281. Rabilloud T, Kieffer S, Procaccio V, Louwagie M, Courch-
polysaccharide from the lipopolysaccharide of Porphy- esne PL, Patterson SD, Martinez P, Garin J, Lunardi J. Two-
romonas gingivalis strain W50. Eur J Biochem 2001: 268: dimensional electrophoresis of human placental mito-
4698±4707. chondria and protein identification by mass spectrometry
267. Parker K, Bednarek M, Coligan J. Scheme for ranking po- ± toward a human mitochondrial proteome. Electrophor-
tential HLA-A2 binding peptides based on independent esis 1998: 19: 1006±1014.
binding of individual peptide side-chains. J Immunol 282. Rajapakse PS, O'Brien-Simpson NM, Slakeski N, Hoffmann
1994: 152: 163±175. B, Reynolds EC. Immunization with the RgpA-Kgp protei-
268. Pederson ED, Miller JW, Matheson S, Simonson LG, Chad- nase±adhesin complexes of Porphyromonas gingivalis pro-
wick DE, Covill PJ, Turner DW, Lamberts BL, Morton HE. tects against periodontal bone loss in the rat periodontitis
Trypsin-like activity levels of Treponema denticola and model. Infect Immun 2002: 70: 2480±2486.
Porphyromonas gingivalls in adults with periodontitis. 283. Rammensee H, Bachmann J, Emmerich N, Bachor O,
J Clin Periodontol 1994: 21: 519±525. Stevanovic S. SYFPEITHI: database for MHC ligands and
269. Perry MB, MacLean LL, Gmur R, Wilson ME. Characteriza- peptide motifs. Immunogenetics 1999: 50: 213±219.
tion of the O-polysaccharide structure of lipopolysacchar- 284. Rangarajan M, Smith SJUS, Curtis MA. Biochemical char-
ide from Actinobacillus actinomycetemcomitans serotype acterization of the arginine-specific proteases of Porphyr-
b. Infect Immun 1996: 64: 1215±1219. omonas gingivalis W50 suggests a common precursor.
270. Perry MB, MacLean LM, Brisson JR, Wilson ME. Structures Biochem J 1997: 323: 701±709.
of the antigenic O-polysaccharides of lipopolysaccharides 285. Reddi K, Poole S, Nair S, Meghji S, Henderson B, Wilson M.
produced by Actinobacillus actinomycetemcomitans sero- Lipid A-associated proteins from periodontopathogenic
types a, c, d and e. Eur J Biochem 1996: 242: 682±688. bacteria induce interleukin-6 production by human

130
 Antigens of bacteria associated with periodontitis

gingival fibroblasts and monocytes. FEMS Immunol Med 301. Schifferle RE, Reddy MS, Zambon JJ, Genco RJ, Levine MJ.
Microbiol 1995: 11: 137±144. Characterization of a polysaccharide antigen from Bacter-
286. Reddi K, Meghji S, Nair SP, Arnett TR, Miller AD, oides gingivalis. J Immunol 1989: 143: 3035±3042.
Preuss M, Wilson M, Henderson B, Hill P. The Escheri- 302. Schultz CP, Wolf V, Lange R, Mertens E, Wecke J, Naumann
chia coli chaperonin 60 (groEL) is a potent stimulator of D, Zahringer U. Evidence for a new type of outer mem-
osteoclast formation. J Bone Miner Res 1998: 13: 1260± brane lipid in oral spirochete Treponema denticola. Func-
1266. tioning permeation barrier without lipopolysaccharides.
287. Reid HI, Riggio MP. Identification and nucleotide sequ- J Biol Chem 1998: 273: 15661±15666.
ence of the heat shock protein 60 (GroEL) gene of Bac- 303. Scott CF, Whitaker EJ, Hammond BF, Colman RW. Purifi-
teroides forsythus. DNA Seq 1998: 9: 359±364. cation and characterization of a potent 70-kDa thiol lysyl-
288. Roper JM, Raux E, Brindley AA, Schubert HL, Gharbia SE, proteinase (Lys-gingivain) from Porphyromonas gingivalis
Shah HN, Warren MJ. The enigma of cobalamin (Vitamin B that cleaves kininogens and fibrinogen. J Biol Chem 1993:
12) biosynthesis in Porphyromonas gingivalis. Identifica- 268: 7935±7942.
tion and characterization of a functional corrin pathway. 304. Seida K, Saito A, Yamada S, Ishihara K, Naito Y, Okuda K. A
J Biol Chem 2000: 275: 40316±40323. sensitive enzymatic method (SK-013) for detection of
289. Rosan B, Slots J, Lamont RJ, Listgarten MA, Nelson GM. Treponema denticola, Porphyromonas gingivals and Bac-
Actinobacillus actinomycetemcomitans fimbriae. Oral Mi- teroides forsythus in subgingival plaque samples. J Perio-
crobiol Immunol 1988: 3: 58±63. dontal Res 1992: 27: 86±91.
290. Rosen G, Sela MN, Naor R, Halabi A, Barak V, Shapira L. 305. Shah HN, Seddon SV, Gharbia SE. Studies on the virulence
Activation of murine macrophages by lipoprotein and li- properties and metabolism of pleiotropic mutants of Por-
pooligosaccharide of Treponema denticola. Infect Immun phyromonas gingivalis (Bacteroides gingivalis) W50. Oral
1999: 67: 1180±1186. Microbiol Immunol 1989: 4: 19±23.
291. Ross BC, Czajkowski L, Hocking D, Margetts M, Webb E, 306. Shapira L, Frolov I, Halabi A, Ben-Nathan D. Experimental
Rothel L, Patterson M, Agius C, Camuglia S, Reynolds E, stress suppresses recruitment of macrophages but en-
Littlejohn T, Gaeta B, Ng A, Kuczek ES, Mattick JS, Gearing hanced their P. gingivalis lipopolysaccharide-stimulated
D, Barr IG. Identification of vaccine candidate antigens secretion of nitric oxide. J Periodontol 2000: 71: 476±481.
from a genomic analysis of Porphyromonas gingivalis. Vac- 307. Sharma A, Sojar HT, Lee JY, Genco RJ. Expression of a
cine 2001: 19: 4135±4142. functional Porphyromonas gingivalis fimbrillin polypep-
292. Roux S, Orcel P. Bone loss. Factors that regulate osteo- tide in Escherichia coli: purification, physicochemical and
clast differentiation: an update. Arthritis Res 2000: 2: immunochemical characterization, and binding character-
451±456. istics. Infect Immun 1993: 61: 3570±3573.
293. Ruby JD, Charon NW. Effect of temperature and viscosity 308. Sharma A, Sojar HT, Glurich I, Honma K, Kuramitsu HK,
on the motility of the spirochete Treponema denticola. Genco RJ. Cloning, expression, and sequencing of a cell
FEMS Microbiol Lett 1998: 169: 251±254. surface antigen containing a leucine-rich repeat motif
294. Ruby JD, Li H, Kuramitsu H, Norris SJ, Goldstein SF, Buttle from Bacteroides forsythus ATCC 43037. Infect Immun
KF, Charon NW. Relationship of Treponema denticola peri- 1998: 66: 5703±5710.
plasmic flagella to irregular cell morphology. J Bacteriol 309. Shenker BJ, Kushner ME, Tsai CC. Inhibition of fibroblast
1997: 179: 1628±1635. proliferation by Actinobacillus actinomycetemcomitans.
295. Saarela M, Asikainen S, Alaluusua S, Pyhala L, Lai CH, Infect Immun 1982: 38: 986±992.
Jousimies-Somer H. Frequency and stability of mono- or 310. Shenker BJ, Listgarten MA, Taichman NS. Suppression of
poly-infection by Actinobacillus actinomycetemcomitans human lymphocyte responses by oral spirochetes: a mono-
serotypes a, b, c, d or e. Oral Microbiol Immunol 1992: 7: cyte-dependent phenomenon. J Immunol 1984: 132:
277±279. 2039±2045.
296. Saito A, Hosaka Y, Nakagawa T, Yamada S, Okuda K. Relative 311. Shenker BJ, Vitale LA, Welham DA. Immune suppression
avidity of serum immunoglobulin G antibody for the induced by Actinobacillus actinomycetemcomitans: effects
fimbria antigen of Actinobacillus actinomycetemcomitans on immunoglobulin production by human B cells. Infect
in patients with adult periodontitis. Infect Immun 1993: 61: Immun 1990: 58: 3856±3862.
332±334. 312. Shenker BJ, Vitale LA, Keiba I, Harrison G, Berthold P,
297. Saito T, Ishihara K, Kato T, Okuda K. Cloning, expression, Golub E, Lally ET. Flow cytometric analysis of the cytotoxic
and sequencing of a protease gene from Bacteroides for- effects of Actinobacillus actinomycetemcomitans leuko-
sythus ATCC 43037 in Escherichia coli. Infect Immun 1997: toxin on human natural killer cells. J Leukoc Biol 1994:
65: 4888±4891. 55: 153±160.
298. Sakamoto M, Suzuki M, Umeda M, Ishikawa L, Benno Y. 313. Shenker BJ, Vitale L, King C. Induction of human T cells
Reclassification of Bacteroides forsythus (Tanner et al. that coexpress CD4 and CD8 by an immunomodulatory
1986) as Tannerella forsythia corrig., gen. nov., comb. protein produced by Actinobacillus actinomycetemcomi-
nov. Int J Syst Evol Microbiol 2002: 52: 841±849. tans. Cell Immunol 1995: 164: 36±46.
299. Sandros J, Papapanou P, Dahlen G. Porphyromonas gingi- 314. Shi Y, Ratnayake DB, Okamoto K, Abe N, Yamamoto K,
valis invades oral epithelial cells in vitro. J Periodontal Res Nakayama K. Genetic analyses of proteolysis, hemoglobin
1993: 28: 219±226. binding, and hemagglutination of Porphyromonas gingiva-
300. Schenck K, Michaelsen TE. IgG subclass distribution of lis. Construction of mutants with a combination of rgpA,
serum antibodies against lipopolysaccharide from Bacter- rgpB, kgp, and hagA. J Biol Chem 1999: 274: 17955±17960.
oides gingivalis in periodontal health and disease. Acta 315. Shibata Y, Kurihara K, Takiguchi H, Abiko Y. Construction
Pathol Microbiol Immunol Scand [C] 1987: 95: 41±46. of a functional single-chain variable fragment antibody

131
O'Brien-Simpson et al.

against hemagglutinin from Porphyromonas gingivalis. In- F, Hammer J. Generation of tissue-specific and promiscu-
fect Immun 1998: 66: 2207±2212. ous HLA ligand databases using DNA microarrays and
316. Shibata Y, Hayakawa M, Takiguchi H, Shiroza T, Abiko Y. virtual HLA class II matrices. Nat Biotechnol 1999: 17:
Determination and characterisation of the hemagglutinin- 555±561.
associated short motifs found in P. gingivalis multiple gene 332. Sugawara S, Nemoto E, Tada H, Miyake K, Imamura T,
products. J Biol Chem 1999: 274: 5012±5020. Takada H. Proteolysis of human monocyte CD14 by cys-
317. Simonson LG, McMahon KT, Childers DW, Morton HE. teine proteinases (gingipains) from Porphyromonas gingi-
Bacterial synergy of Treponema denticola and Porphyro- valis leading to lipopolysaccharide hyporesponsiveness.
monas gingivalis in a multinational population. Oral Mi- J Immunol 2000: 165: 411±418.
crobiol Inummol 1992: 7: 111±112. 333. Sugita N, Kimura A, Matsuki Y, Yamamoto T, Yoshie H,
318. Simpson DL, Berthold P, Taichman NS. Killing of human Hara K. Activation of transcription factors and IL-8 expres-
myelomonocytic leukemia and lymphocytic cell lines by sion in neutrophils stimulated with lipopolysaccharide
Actinobacillus actinomycetemcomitans leukotoxin. Infect from Porphyromonas gingivalis. Inflammation 1998: 22:
Immun 1988: 56: 1162±1166. 253±267.
319. Sims TJ, Moncla BJ, Darveau RP, Page RC. Antigens of 334. Sundqvist G, Bengtson A, Carlsson J. Generation and de-
Actinobacillus actinomycetemcomitans recognized by pa- gradation of the complement fragment C5a in human se-
tients with juvenile periodontitis and periodontally normal rum by Bacteroides gingivalis. Oral Microbiol Immunol
subjects. Infect Immun 1991: 59: 913±924. 1988: 3: 103±107.
320. Sims TJ, Mancl LA, Braham PH, Bainbridge BW, Page RC. 335. Sundqvist G, Figdor D, HaÈnstroÈm L, SoÈrlin S, SandstroÈm G.
Antigenic variation and cross-reactivity in Bacteroides for- Phagocytosis and virulence of different strains of Porphyr-
sythus clinical isolates detected by western blot. J Clin omonas gingivalis. Scand J Dent Res 1991: 99: 117±129.
Immunol 1998: 18: 355±367. 336. Tabeta K, Yamazaki K, Akashi S, Miyake K, Kumada H,
321. Slakeski N, Cleal SM, Reynolds EC. Characterization of a Umemoto T, Yoshie H. Toll-like receptors confer respon-
Porphyromonas gingivalis gene prtR that encodes an argi- siveness to lipopolysaccharide from Porphyromonas gingi-
nine-specific thiol proteinase and multiple adhesins. Bio- valis in human gingival fibroblasts. Infect Immun 2000: 68:
chem Biophys Res Comm 1996: 224: 605±610. 3731±3735.
322. Slakeski N, Bhogal PS, O'Brien-Simpson NM, Reynolds EC. 337. Tabeta K, Yamazaki K, Hotokezaka H, Yoshie H, Hara K.
Characterisation of a second cell-associated Arg-specific Elevated humoral immune response to heat shock protein
cysteine proteinase of Porphyromonas gingivalis and iden- 60 (hsp60) family in periodontitis patients. Clin Exp Im-
tification of an adhesin binding motif involved in associa- munol 2000: 120: 285±293.
tion of the PrtR and PrtK proteinases and adhesins into 338. Tabeta K, Yoshie H, Yamazaki K. Characterization of serum
large complexes. Microbiology 1998: 144: 1583±1592. antibody to Actinobacillus actinomycetemcomitans GroEL-
323. Slakeski N, Cleal SM, Bhogal PS, Reynolds EC. Character- like protein in periodontitis patients and healthy subjects.
ization of a Porphyromonas gingivalis gene prtK that Oral Microbiol Immunol 2001: 16: 290±295.
encodes a lysine-specific cysteine proteinase and three 339. Tada H, Sugawara S, Nemoto E, Takahashi N, Imamura T,
sequence-related adhesins. Oral Microbiol Immunol Potempa J, Travis J, Shimauchi H, Takada H. Proteolysis of
1999: 14: 92±97. CD14 on human gingival fibroblasts by arginine-specific
324. Slakeski N, Dashper SG, Cook P, Poon C, Moore C, cysteine proteinases from Porphyromonas gingivalis lead-
Reynolds EC. A Porphyromonas gingivalis genetic locus ing to down-regulation of lipopolysaccharide-induced in-
encoding a heme transport system. Oral Microbiol Immunol terleukin-8 production. Infect Immun 2002: 70: 3304±3307.
2000: 15: 388±391. 340. Taichman NS, Wilton JM. Leukotoxicity of an extract
325. Slots J. The predominant cultivable organisms in juvenile from Actinobacillus actinomycetemcomitans for human
periodontitis. Scand J Dent Res 1976: 84: 1±10. gingival polymorphonuclear leukocytes. Inflammation
326. Smalley JW, Birss AJ, Kay HM, McKee AS, Marsh PD. The 1981: 5: 1±12.
distribution of trypsin-like enzyme activity in cultures of a 341. Taichman NS, Iwase M, Korchak H, Berthold P, Lally ET.
virulent and an avirulent strain of Bacteroides gingivalis Membranolytic activity of Actinobacillus actinomycetemco-
W50. Oral Microbiol Immunol 1989: 4: 178±181. mitans leukotoxin. J Periodontal Res 1991: 26: 258±260.
327. Socransky SS, Haffajee AD. The bacterial eitiology of de- 342. Takada H, Hirai H, Fujiwara T, Koga T, Ogawa T, Hamada
structive periodontal disease: current concepts. J Period- S. Bacteroides lipopolysaccharides (LPS) induce anaphy-
ontol 1992: 63: 322±331. lactoid and lethal reactions in lipopolysaccharide-respon-
328. Socransky SS, Haffajee AD, Cugini MA, Smith C, Kent RL, sive and -nonresponsive mice primed with muramyl
Jr. Microbial complexes in subgingival plaque. J Clin Per- dipeptide. J Infect Dis 1990: 162: 428±434.
iodontol 1998: 25: 134±144. 343. Takahashi T, Nishihara T, Ishihara Y, Amano K, Shibuya N,
329. Sojar HT, Sharma A, Genco RJ. Porphyromonas gingivalis Moro I, Koga T. Murine macrophage interleukin-1 release
fimbriae bind to cytokeratin of epithelial cells. Infect Im- by capsularlike serotype-specific polysaccharide antigens
mun 2002: 70: 96±101. of Actinobacillus actinomycetemcomitans. Infect Immun
330. Sorsa T, Ingman T, Suomalainen K, Haapasalo M, Kontti- 1991: 59: 18±23.
nen YT, Lindy O, Saari H, Uitto VJ. Identification of pro- 344. Takeshita A, Murakami Y, Yamashita Y, Ishida M, Fujisawa
teases from periodontopathogenic bacteria as activators of S, Kitano S, Hanazawa S. Porphyromonas gingivalis fim-
latent human neutrophil and fibroblast-type interstitial briae use beta2 integrin (CD11/CD18) on mouse peritoneal
collagenases. Infect Immun 1992: 60: 4491±4495. macrophages as a cellular receptor, and the CD18 beta
331. Sturniolo T, Bono E, Ding J, Raddrizzani L, Tuereci O, chain plays a functional role in fimbrial signaling. Infect
Sahin U, Braxenthaler M, Gallazzi F, Protti M, Sinigaglia Immun 1998: 66: 4056±4060.

132
 Antigens of bacteria associated with periodontitis

345. Tamura M, Tokuda M, Nagaoka S, Takada H. Lipopolysac- 362. Van der Zee R, Van Eden W, Meloen R, Noordzij A, Van
charides of Bacteroides intermedius (Prevotella intermedia) Embden J. Efficient mapping and characterization of a T
and Bacteroides (Porphyromonas) gingivalis induce inter- cell epitope by the simultaneous synthesis of multiple pep-
leukin-8 gene expression in human gingival fibroblast cul- tides. Eur J Immunol 1989: 19: 43±47.
tures. Infect Immun 1992: 60: 4932±4937. 363. Van Steenbergen TJ, Kastelein P, Touw JJ, de Graaff J.
346. Tan KS, Song KP, Ong G. Bacteroides forsythus prtH geno- Virulence of black-pigmented Bacteroides strains from pe-
type in periodontitis patients: occurrence and association riodontal pockets and other sites in experimentally induced
with periodontal disease. J Periodontal Res 2001: 36: skin lesions in mice. J Periodontal Res 1982: 17: 41±49.
398±403. 364. Van Steenbergen TJ, Delemarre FG, Namavar F, De Graaff
347. Tanamoto K, Azumi S, Haishima Y, Kumada H, Umemoto J. Differences in virulence within the species Bacteroides
T. Endotoxic properties of free lipid A from Porphyromonas gingivalis. Antonie Van Leeuwenhoek 1987: 53: 233±244.
gingivalis. Microbiology 1997: 143: 63±71. 365. Vasel D, Sims TJ, Bainbridge B, Houston L, Darveau R,
348. Tani Y, Tani M, Kato I. Extracellular 37-kDa antigenic pro- Page RC. Shared antigens of Porphyromonas gingivalis
tein from Actinobacillus actinomycetemcomitans induces and Bacteroides forsythus. Oral Microbiol Immunol 1996:
TNF-alpha, IL-1 beta, and IL-6 in murine macrophages. 11: 226±235.
J Dent Res 1997: 76: 1538±1547. 366. Veith PD, Talbo GH, Slakeski N, Reynolds EC. Identifica-
349. Tanner A, Stillman N. Oral and dental infections with tion of a novel heterodimeric outer membrane protein of
anaerobic bacteria: clinical features, predominant patho- Porphyromonas gingivalis by two-imensional gel electro-
gens, and treatment. Clin Infect Dis 1993: 16: S304±S309. phoresis and peptide mass fingerprinting. Eur J Biochem
350. Tanner AC, Haffer C, Bratthall GT, Visconti RA, Socransky 2001: 268: 4748±4757.
SS. A study of the bacteria associated with advancing peri- 367. Veith PD, Talbo GH, Slakeski N, Dashper SG, Moore C,
odontitis in man. J Clin Periodontol 1979: 6: 278±307. Paolini RA, Reynolds EC. Major outer membrane proteins
351. Tanner AC, Strzempko MN, Belsky CA, McKinley GA. API and proteolytic processing of RgpA and Kgp of Porphyro-
ZYM and API An-Ident reactions of fastidious oral gram- monas gingivalis W50. Biochem J 2002: 363: 105±115.
negative species. J Clin Microbiol 1985: 22: 333±335. 368. Wang PL, Shinohara M, Murakawa N, Endo M, Sakata S,
352. Taubman MA, Yoshie H, Wetherell JR Jr, Ebersole JL, Smith Okamura M, Ohura K. Effect of cysteine protease of Por-
DJ. Immune response and periodontal bone loss in germ- phyromonas gingivalis on adhesion molecules in gingival
free rats immunized and infected with Actinobacillus acti- epithelial cells. Jpn J Pharmacol 1999: 80: 75±79.
nomycetemcomitans. J Periodontal Res 1983: 18: 393±401. 369. Wang PL, Shirasu S, Shinohara M, Daito M, Fujii T, Kowa-
353. Tokuda M, Duncan M, Cho M-I, Kuramitsu HK. Role of shi Y, Ohura K. Purification and characterization of a tryp-
Porphyromonas gingivalis protease activity in colonization sin-like protease from the culture supernatant of
of oral surfaces. Infect Immun 1996: 64: 4067±4073. Actinobacillus actinomycetemcomitans Y4. Eur J Oral Sci
354. Tokuda M, Karunakaran T, Duncan M, Hamada N, Kura- 1999: 107: 147±153.
mitsu H. Role of Arg-gingipain A in virulence of Porphy- 370. Wang PL, Azuma Y, Shinohara M, Ohura K. Effect of
romonas gingivalis. Infect Immun 1998: 66: 1159±1166. Actinobacillus actinomycetemcomitans protease on the
355. Ueda N, Koide M, Ohguchi M, Ishihara Y, Noguchi T, proliferation of gingival epithelial cells. Oral Dis 2001: 7:
Okahashi N, Nishihara T. Involvement of prostaglandin 233±237.
E2 and interleukin-1 alpha in the differentiation and sur- 371. Wang PL, Oido-Mori M, Fujii T, Kowashi Y, Kikuchi M,
vival of osteoclasts induced by lipopolysaccharide from Suetsugu Y, Tanaka J, Azuma Y, Shinohara M, Ohura K.
Actinobacillus actinomycetemcomitans Y4. J Periodontal Heterogeneous expression of Toll-like receptor 4 and
Res 1998: 33: 509±516. downregulation of Toll-like receptor 4 expression on hu-
356. Ueki K, Tabeta K, Yoshie H, Yamazaki K. Self-heat shock man gingival fibroblasts by Porphyromonas gingivalis lipo-
protein 60 induces tumour necrosis factor alpha in mono- polysaccharide. Biochem Biophys Res Commun 2001: 288:
cyte-derived macrophage: possible role in chronic inflam- 863±867.
matory periodontal disease. Clin Exp Immunol 2002: 127: 372. Watanabe H, Marsh PD, Ivanyi L. Antigens of Actinobacil-
72±77. lus actinomycetemcomitans identified by immunoblotting
357. Uitto VJ, Grenier D, Chan EC, McBride BC. Isolation of a with sera from patients with localized human juvenile
chymotrypsinlike enzyme from Treponema denticola. In- periodontitis and generalized severe periodontitis. Arch
fect Immun 1988: 56: 2717±2722. Oral Biol 1989: 34: 649±656.
358. Uitto VJ, Pan YM, Leung WK, Larjava H, Ellen RP, Finlay 373. Watanabe H, Marsh PD, Ivanyi L. Detection of immuno-
BB, McBride BC. Cytopathic effects of Treponema dentico- dominant antigens of periodontopathic bacteria in human
la chymotrypsin-like proteinase on migrating and strati- periodontal disease. Oral Microbiol Immunol 1989: 4:
fied epithelial cells. Infect Immun 1995: 63: 3401±3410. 159±164.
359. Umemoto T, Namikawa I, Suido H, Asai S. A major antigen 374. Watanabe K, Yamaji Y, Umemoto T. Correlation between
on the outer envelope of a human oral spirochete, Trepo- cell-adherent activity and surface structure in Porphyro-
nema denticola. Infect Immun 1989: 57: 2470±2474. monas gingivalis. Oral Microbiol Immunol 1992: 7: 357±
360. Umemoto T, Nakatani Y, Nakamura Y, Namikawa I. Fibro- 363.
nectin-binding proteins of a human oral spirochete Trepo- 375. Weinberg A, Holt SC. Interaction of Treponema denticola
nema denticola. Microbiol Immunol 1993: 37: 75±78. TD-4, GM-1, and MS25 with human gingival fibroblasts.
361. Umemoto T, Wadood A, Nakamura Y, Nakatani Y, Nami- Infect Immun 1990: 58: 1720±1729.
kawa I. Antigenic behaviors of two axial flagellar proteins 376. Weinberg A, Holt SC. Chemical and biological activities of
detected in Treponema denticola. Microbiol Immunol 1993: a 64-kilodalton outer sheath protein from Treponema den-
37: 159±163. ticola strains. J Bacteriol 1991: 173: 6935±6947.

133
O'Brien-Simpson et al.

377. Weinberg A, Belton CM, Park Y, Lamont RJ. Role of fim- 392. Wyss C. Dependence of proliferation of Bacteroides for-
briae in Porphyromonas gingivalis invasion of gingival sythus on exogenous N-acetylmuramic acid. Infect Immun
epithelial cells. Infect Immun 1997: 65: 313±316. 1989: 57: 1757±1759.
378. Weldingh K, Rosenkrands I, Jacobsen S, Rasmussen PB, 393. Yamashita K, Eastcott JW, Taubman MA, Smith DJ, Cox DS.
Elhay MJ, Andersen P. Two-dimensional electrophoresis Effect of adoptive transfer of cloned Actinobacillus actino-
for analysis of Mycobacterium tuberculosis culture filtrate mycetemcomitans-specific T helper cells on periodontal
and purification and characterization of six novel proteins. disease. Infect Immun 1991: 59: 1529±1534.
Infect Immun 1998: 66: 3492±3500. 394. Yamazaki K, Ikarashi F, Aoyagi T, Takahashi K, Nakajima T,
379. Welling GW, Weijer WJ, van Der Zee R, Welling-Webster S. Hara K, Seymour GJ. Direct and indirect effects of Porphy-
Prediction of sequential antigenic reions in proteins. FEBS romonas gingivalis lipopolysaccharide on interleukin-6
Letters 1985: 188: 215. production by human gingival fibroblasts. Oral Microbiol
380. Wendell KJ, Stein SH. Regulation of cytokine production Immunol 1992: 7: 218±224.
in human gingival fibroblasts following treatment with 395. Yamazaki K, Ohsawa Y, Tabeta K, Ito H, Ueki K, Oda T,
nicotine and lipopolysaccharide. J Periodontol 2001: 72: Yoshie H, Seymour GJ. Accumulation of human heat shock
1038±1044. protein 60-reactive T cells in the gingival tissues of period-
381. White PA, Wilson M, Nair SP, Kirby AC, Reddi K, Hender- ontitis patients. Infect Immun 2002: 70: 2492±2501.
son B. Characterization of an antiproliferative surface- 396. Yanagita M, Hiroi T, Kitagaki N, Hamada S, Ito HO, Shi-
associated protein from Actinobacillus actinomycetemco- mauchi H, Murakami S, Okada H, Kiyono H. Nasopharyn-
mitans which can be neutralized by sera from a proportion geal-associated lymphoreticular tissue (NALT) immunity:
of patients with localized juvenile periodontitis. Infect fimbriae-specific Th1 and Th2 cell-regulated IgA responses
Immun 1995: 63: 2612±2618. for the inhibition of bacterial attachment to epithelial cells
382. Wick G, Perschinka H, Millonig G. Atherosclerosis as an and subsequent inflammatory cytokine production. J Im-
autoimmune disease: an update. Trends Immunol 2001: 22: munol 1999: 162: 3559±3565.
665±669. 397. Yoneda M, Hirofuji T, Anan H, Matsumoto A, Hamachi T,
383. Wilson ME. IgG antibody response of localized juvenile Nakayama K, Maeda K. Mixed infection of Porphyromonas
periodontitis patients to the 29 kilodalton outer membrane gingivalis and Bacteroides forsythus in a murine abscess
protein of Actinobacillus actinomycetemcomitans. J Perio- model: involvement of gingipains in a synergistic effect.
dontol 1991: 62: 211±218. J Periodontal Res 2001: 36: 237±243.
384. Wilson ME, Hamilton RG. Immunoglobulin G subclass 398. Yoshimura F, Takahashi K, Nodasaka Y, Suzuki T. Purifica-
response of localized juvenile periodontitis patients to Ac- tion and characterization of a novel type of fimbriae from
tinobacillus actinomycetemcomitans Y4 lipopolysacchar- the oral anaerobe Bacteroides gingivalis. J Bacteriol 1984:
ide. Infect Immun 1992: 60: 1806±1812. 160: 949±957.
385. Wilson M, Henderson B. Virulence factors of Actinobacillus 399. Yoshimura F, Sugano T, Kawanami M, Kato H, Suzuki T.
actinomycetemcomitans relevant to the pathogenesis of Detection of specific antibodies against fimbriae and
inflammatory periodontal diseases. FEMS Microbiol Rev membrane proteins from the oral anaerobe Bacteroides
1995: 17: 365±379. gingivalis in patients with periodontal diseases. Microbiol
386. Wilson ME, Hamilton RG. Immunoglobulin G subclass Immunol 1987: 31: 935±941.
response of juvenile periodontitis subjects to principal 400. Yoshimura A, Hara Y, Kaneko T, Kato I. Secretion of IL-1
outer membrane proteins of Actinobacillus actinomyce- beta, TNF-alpha, IL-8 and IL-1ra by human polymorpho-
temcomitans. Infect Immun 1995: 63: 1062±1069. nuclear leukocytes in response to lipopolysaccharides
387. Wingrove JA, DiScipio RG, Chen Z, Potempa J, Travis J, from periodontopathic bacteria. J Periodontal Res 1997:
Hugh TE. Activation of complement components C3 and 32: 279±286.
C5 by a cysteine proteinase (gingipain-1) from Porphyro- 401. Yotis WW, Macaluso F, Gopalsami C. Immunochemical
monas (Bacteroides) gingivalis. J Biol Chem 1992: 267: features of a macromolecule of Treponema denticola. J
18902±18907. Basic Microbiol 1995: 35: 255±268.
388. van Winkelhoff AJ, Appelmelk BJ, Kippuw N, de Graaff J. K- 402. Zambon JJ. Periodontal diseases: microbial factors. Ann
antigens in Porphyromonas gingivalis are associated with Periodontol 1996: 1: 879±925.
virulence. Oral Microbiol Immunol 1993: 8: 259±265. 403. Zambon J, DeLuca C, Slots J, Genco R. Studies of leukotoxin
389. Wolf V, Lange R, Wecke J. Development of quasi-multi- from Actinobacillus actinomycetemcomitans using the pro-
cellular bodies of Treponema denticola. Arch Microbiol myelocytic HL-60 cell line. Infect Immun 1983: 40: 205±212.
1993: 160: 206±213. 404. Zambon JJ, Slots J, Genco RJ. Serology of oral Actinobacil-
390. Woo DD, Holt SC, Leadbetter ER. Ultrastructure of Bac- lus actinomycetemcomitans and serotype distribution in
teroides species: Bacteroides asaccharolyticus, Bacteroides human periodontal disease. Infect Immun 1983: 41: 19±27.
fragilis, Bacteroides melaninogenicus subspecies melanino- 405. Zhao S, Shi J, Xiao M. Effects of capsules on attachment of
genicus, and B. melaninogenicus subspecies intermedius. human periodontal ligament fibroblasts to root surface.
J Infect Dis 1979: 139: 534±546. Zhonghua Kou Qiang Yi Xue Za Zhi 1996: 31: 19±21.
391. Wray D, Grahame L. Periodontal bone loss in mice induced 406. Zugel U, Kaufmann SH. Role of heat shock proteins in
by different periodontopathic organisms. Arch Oral Biol protection from and pathogenesis of infectious diseases.
1992: 37: 435±438. Clin Microbiol Rev 1999: 12: 19±39.

134