molecular oral
microbiology
molecular oral microbiology
Dental caries – not just holes in teeth!
A perspective
W.H. Bowen
School of Medicine and Dentistry, Center for Oral Biology, University of Rochester Medical Center, Rochester, NY, USA
Correspondence: William H. Bowen, School of Medicine and Dentistry, Center for Oral Biology, University of Rochester Medical Center, 601
Elmwood Avenue, Box 611, Rochester, NY 14642, USA Tel.: +1 585 275 0772; fax: +1 585 276 0190; E-mail: william_bowen@urmc.
rochester.edu
Keywords: biofilms; dental caries; diffusion; glucosyltransferase; matrix; saliva
Accepted 3 September 2015
DOI: 10.1111/omi.12132
SUMMARY
Cavitation in teeth results from a pathogenic pro-
cess termed dental caries that has occurred on
the tooth surface for weeks or even years. Accu-
mulation of dental plaque (biofilm) on the tooth is
usually the first manifestation of the disease.
Although acid production is the immediate and
proximal cause of dissolution of teeth; it is the
milieu within which the acid is formed that should
be of primary concern. Focusing on the ‘critical
pH’ has detracted attention from the more biologi-
cal aspects (biofilm formation) of dental caries.
Dental caries is unique; it is a biological process
occurring on essentially an inert surface. Investi-
gation of the multitude of interactions occurring
in plaque ranging from enamel interfaces to sur-
faces of bacteria and matrices poses challenges
worthy of the best scientific minds. The mouth
clearly offers unique opportunities to investigate
the multi facets of biofilm formation in vivo, gen-
erating data that have relevance way beyond the
mouth. Prevention of this ubiquitous disease,
dental caries, continues to present serious
challenges. The public health benefits of fluoride
delivered in its various formats are well
recognized. Nevertheless, additional preventive
approaches are required. Overcoming the rapid
clearance of agents from the mouth is particularly
challenging. Building on the polymerizing capac-
ity of glucosyltransferases it may be possible to
incorporate a therapeutic agent into the matrix
228
INVITED REVIEW
plaque, thereby delivering therapeutic agents pre-
cisely to where they are needed.
INTRODUCTION
The term ‘Dental Caries’ first appeared in the litera-
ture around 1634 according to the Merriam medical
dictionary. It is derived from the Latin to decay and
from ancient Irish Ara Chrinn meaning it decays or
withers. The term was originally used simply to
describe holes in the teeth with little knowledge of the
etiology and pathogenesis of the disease. As a result,
the word ‘caries’ is used to describe lesions at vari-
ous stages of development. In addition, it is used in
the context that ‘caries’ was diagnosed and even
treated. Misuse is unfortunate because it has led to
imprecision and almost a strictly limited focus on cari-
ous lesions. The lesion is an overt expression of a
pathologic process that has been occurring on the
tooth surface for months or even years. As Appleton
(1944) points out, ‘it is well to distinguish between a
process and the result of this process between caries
and the resulting cavity’. Most dentists by reason of
their training focus on restoring teeth but refer to the
procedure as treating dental caries. Inserting restora-
tions has little or no influence on the pathogenesis of
the disease ‘dental caries’. The term ‘dental caries’
should be restricted to the actual disease: cavities
could then be referred to as carious lesions. Dental
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
W.H. Bowen
caries is often ‘diagnosed’ by means of a probe or
radiographs.
Research into the etiology and pathogenesis of
dental caries has probably been hampered by focus-
ing on the carious lesion to the virtual exclusion of
exploring the pathophysiology of the disease. Much
attention has been given to acid dissolution of the
tooth and the so-called critical pH (Bowen, 2013a).
For numerous decades investigators have focused on
neutralizing acid to the virtual exclusion of all other
avenues as a means of preventing the development
of carious lesions. Such an approach appears to be
oblivious to the fact that there is more than sufficient
buffering power through saliva to neutralize any acid
produced in the mouth. So why does neutralization
not occur? Why do lesions develop in the alkaline
mouth? Put simply, saliva cannot gain access to the
acid produced in the depths of plaque that has accu-
mulated on tooth surfaces. Plaque is of course a bio-
film, and all biofilms by definition have a matrix
(Flemming & Wingender, 2010). The chemical and
physical properties of the matrix hinder the diffusion
of bicarbonate and other potentially therapeutic sub-
stances into plaque (Bowen & Koo, 2011).
The matrix of dental plaque is composed primarily
of glucan and, to a lesser extent, fructan (Bowen &
Koo, 2011). This observation has been repeated in
many animals fed sucrose. The exact composition
and structure of the glucan is influenced, in part at
least, by the interval since the last intake of sucrose
and starch. In general, the glucan in mature plaque is
insoluble, and is mainly composed of a1-3-, a1-4-,
a1-6-linked glucose. There is some evidence that it is
charged through binding of phosphates, possibly
lipoteichoic acid, and DNA (Liao et al., 2014). Hence,
it is ideally suited to block access by saliva through
limiting diffusion physically and by charge. However,
the glucan and the glucosyltransferases (Gtfs) that
produce it contribute to the potential virulence of pla-
que biofilm in additional ways. For example, the
enzymes GtfB, GtfC, and GtfD, produced by strepto-
cocci, adhere strongly to saliva-coated tooth surfaces
and to other biological surfaces such as bacteria.
Enzyme C has the greatest affinity for saliva-coated
surfaces and remains active on the surface; following
exposure to sucrose the surface-formed glucan pro-
vides novel binding sites on teeth to which many
organisms may adhere (Vacca-Smith & Bowen,
1998). It is perhaps surprising that many more
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
Dental caries – a perspective
organisms adhere to a glucan surface than to naked
or saliva-coated hydroxyapatite surfaces. Furthermore
the B enzyme adheres avidly to the surfaces of other
oral microorganisms thereby converting them into de
facto glucan producers when exposed to sucrose.
This phenomenon explains the observation from
transmission electron micrographs, which reveal
microorganisms enmeshed in a polysaccharide matrix
(Bowen & Koo, 2011). Recently it has been observed
that GtfB adheres strongly to the fungus Candida and
apparently facilitates the colonization of plaque by
Candida in the presence of sucrose (Falsetta et al.,
2014). The virulence of the resulting plaque, as
observed in an animal model, is greatly enhanced
through increased formation of insoluble polysaccha-
ride (Falsetta et al., 2014) illustrating, yet again, the
critical role played by the matrix of plaque in the
pathogenesis of dental caries. It is noteworthy that
starch hydrolysates may serve as acceptors in the
presence of sucrose for GtfB thereby adding to the
complexity of structure and perhaps offering an expla-
nation, in part at least, as to why combinations of
starch and sucrose are much more cariogenic than
either one alone (Bowen et al., 1980; Vacca-Smith
et al., 1996). It is conceivable that additional oligosac-
charides may act as acceptors.
Results from several studies reveal that the chemi-
cal structure of a matrix is also influenced by addi-
tional factors (Bowen & Koo, 2011). Immediately
following exposure to sucrose, a1-6 glucan and fruc-
tan are dominant. As plaque matures, the level of
fructan declines and a1-3, a1-4, and a3-4 glucans,
which are insoluble, increase. These changes in com-
position may be explained in part by several phenom-
ena that occur concurrently. The fructan and soluble
glucan are metabolized by bacteria within the matrix,
thereby contributing to acidogenicity. The soluble glu-
can may also serve to enhance activity of GtfB and
serve as a template or primer for complex glucan pro-
duction. Furthermore, dextranase and mutanase, also
produced by oral microorganisms, serve to alter the
structure of glucan, rendering it more insoluble and
enhancing its diffusion-limiting properties. Hence, the
structure of the matrix is not static and undergoes
modifications continuously (Bowen & Koo, 2011).
Clearly the pathophysiology of biofilms is worthy of
expanded exploration. Much of the milieu within den-
tal plaque is highly acidogenic following exposure to
sugars. As expected, the majority of microorganisms
229
Dental caries – a perspective
in plaque are acidogenic and aciduric; frequently
though, Streptococcus mutans is in the minority
(Svensa €ter et al., 2003). This of course is contrary to
what some investigators have come to expect, in par-
ticular if acid is to the forefront in their thinking of the
pathogenesis of dental caries. However, the primary
role of S. mutans in the etiology of dental caries
appears to be providing the matrix of biofilms, thereby
protecting the acidic milieus within which a plethora
of acidogenic organisms thrive. It has been noted in
other biofilms in the body that the organism that initi-
ates formation is frequently found in low numbers and
even occasionally totally absent in mature biofilms
(Flemming & Wingender, 2010). Furthermore the dif-
ferent glucans present in the matrix of plaque offer
binding sites for additional organisms thereby adding
to the microbial complexity in dental plaque. Hence
acid is probably not the sole selective agent for
microorganisms found in ‘mature’ plaque (Vacca-
Smith & Bowen, 1998).
Although many organisms within dental plaque
have the capacity to form acid rapidly, relatively few
can form insoluble glucan from sucrose. Streptococ-
cus mutans is the prime exponent of being an initiator
of plaque and a potent producer of acid. Hence, it
appears that perhaps the role of S. mutans in the eti-
ology and pathogenesis of dental caries needs to be
carefully re-examined. Such a re-evaluation might
prompt more research into avenues to prevent or
remove plaque. It may perhaps suggest ways to deli-
ver an agent to plaque by using the synthesizing
capacity of GtfB enzymes to incorporate therapeutic
substances into plaque and perhaps also other bio-
films (see below). Many plaque biofilms also have the
capacity to form alkali. It is not uncommon to observe
that the pH value of plaque not exposed to sugars
overnight may be more alkaline than the surrounding
saliva. Although recently enhanced attention has
been given to alkali-generating pathways in plaque
and the microorganisms responsible, exploration is
sparse compared with attention devoted to acid pro-
duction (Bowen, 2013a).
CLINICAL RESEARCH: MARKERS
Conducting caries-based research in humans can be
fraught. There are numerous studies claiming rela-
tionship between some current phenomenon and the
prevalence of carious lesions. These studies are
230
W.H. Bowen
frequently conducted in an effort to identify a clinical
marker to aid in diagnosing persons who have the
disease dental caries before the appearance of
lesions. It is extraordinarily difficult to draw clinically
relevant conclusions from currently observed phe-
nomena and the total accumulation of carious lesions.
In addition, in some clinical studies, subjects are
reported as being ‘caries inactive’ because they have
had restorations placed in their teeth. Evidence that
placing restorations alone affects the pathogenesis of
caries is scarce indeed.
Counts of S. mutans in saliva are not infrequently
used as a marker for susceptibility to the disease.
There are several potential difficulties using this
approach. Highly selective microbial growth media
are usually employed, based on the assumption that
all mutans streptococci grow equally well on selective
media. Furthermore, mutans streptococci found in sal-
iva are mostly derived from plaque, and are shed in
clumps and provide sparse information on the sites of
teeth infected. There appears to be little consistency
in the manner by which the saliva is dispersed. It is
hardly surprising therefore that data obtained are also
inconsistent although overall there appears to be a
correlation between the population of S. mutans and
the prevalence of caries in study populations (Tanzer
et al., 2001). Data from such studies could perhaps
be enhanced by the introduction of standardized tech-
niques used consistently. Nevertheless it is important
to point out that many excellent clinical studies of
therapeutic agents have been conducted. These stud-
ies are expensive, difficult and long lasting. If a sim-
ple, reliable biomarker or surrogate could be
identified, perhaps many more agents might be inves-
tigated clinically, and furthermore the way might be
opened for expanded clinical research.
MODELS
Development of novel concepts, to aid in understand-
ing the pathogenesis of dental caries, frequently
entails the use of models. It is essential to understand
that they are just models and that all models suffer
from deficiencies: the model is not the disease, and
that there is a danger of over-interpreting data. It is
disheartening, for example, to see substances that
prevent demineralization in an in vitro model being
described as preventing dental caries – even in the
paper title!
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
W.H. Bowen
Models of plaque biofilm appear in abundance.
Many have properties in common, e.g. convenience
and rapid accumulation of data. Sparse attention, all
too frequently, appears to be given to inclusion of sal-
iva and most importantly to the formation and struc-
ture of matrix. The substrates used vary from glass,
plastic, hydroxyapatite, bovine enamel, and enamel
from humans, all offering distinct surface properties.
Sucrose is frequently omitted, and the resulting
matrix, if any, does not resemble those found in pla-
que from humans. It appears improbable that data
derived from such models provide an adequate plat-
form from which to launch clinical studies.
The use of animal models has contributed enor-
mously to our understanding of the etiology and
pathogenesis of dental caries. Furthermore, data from
animal-based studies have facilitated the develop-
ment of preventive programs. Although the use of ani-
mals in caries-related research has declined recently,
their use continues to have great utility. An outstand-
ing feature of an animal model is of course that the
full spectrum of the disease including role of diet,
micro-organisms, tooth composition, de-mineralization
and re-mineralization, and their interactions may be
explored. Conducting an animal study is not a trivial
exercise (Bowen, 2013b).
Details to be attended to include, for example,
using an adequate number of animals of an appropri-
ate age to ensure a statistically valid result. All too
frequently studies are not commenced until suscepti-
bility to disease has dropped dramatically, usually
around 27 days in rodents (Bowen, 2013b). The addi-
tion of potential therapeutic agents to diet may affect
the eating patterns of the animals, thereby rendering
interpretation of results exceptionally difficult. Data
from animal studies, in contrast to some other mod-
els, tend to arrive slowly; animal studies (rodent) may
take 5–6 months to plan, execute and evaluate
depending on complexity of design and number of
animals included.
PREVENTION
Prevention of the disease dental caries is indeed
challenging. It is well recognized that 80% of the dis-
ease occurs in 20% of the population. Much of the
disease occurs in the economically deprived who can
ill-afford the cost of conventional oral hygiene prod-
ucts. Clearly fluoride in its various forms is of critical
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
Dental caries – a perspective
importance in the prevention of cavities. The ambient
levels of fluoride in the mouth play a major role in the
effectiveness of fluoride; difficulty arises however in
increasing the therapeutic levels in the mouth without
enhancing the risks of fluorosis. Fluoride used in a
conventional way, similar to most topical agents, is
cleared from the mouth relatively quickly, before it
has time to exert its maximum therapeutic potential.
Use of fluoride-releasing devices has been shown to
be highly effective in preventing lesions in small stud-
ies (Meyerowitz & Watson, 1998); the principle is cer-
tainly worthy of additional exploration.
Although dental caries has been recognized for
many decades as an infectious and transmittable dis-
ease (Tanzer et al., 2001) it appears that little effort
has been expended to use this information clinically.
Available evidence reveals that suppression of the
S. mutans population in mothers reduces the level of
infection in children and furthermore that the effect
can be detected years later, as revealed by a
reduced level of decay (Ko €hler & Andre en, 2012).
This observation emphasizes again the importance of
prenatal dental care, that dental caries is a family dis-
ease, and recognizes clinical opportunities to prevent
dental caries.
Although there is a myriad of agents capable of
suppressing the oral microflora, in general none
appears capable of selectively inhibiting S. mutans
specifically. It is important to note that potential thera-
peutic agents that are spectacularly successful
in vitro, killing planktonic cells or inhibiting enzymes
in solution, lack effect when tested in biofilms or
against insolubilized enzymes (Bowen & Koo, 2011).
Recently, attention has been focused on agents that
may disrupt or prevent formation of biofilms through
inhibiting the Gtf enzymes (Bowen & Koo, 2011). All
agents currently used for prevention or treatment of
dental caries have one or more problems. Most are
cleared from the mouth so rapidly that they cannot
exert their maximum therapeutic potential. Further-
more, few agents can penetrate the matrix or effec-
tively remain on the surface of plaque (Bowen & Koo,
2011; Robinson, 2011). Hence, the effective delivery
of potentially therapeutic agents is difficult and
challenging.
It is perhaps possible to use the synthesizing
capacity of plaque to deliver therapeutic agents to
plaque. It is well recognized (Vacca-Smith et al.,
1996) that Gtfs can use acceptors such as
231
Dental caries – a perspective
low-molecular-weight dextrans, and starch hydroly-
sates, which are then incorporated into the full glucan
molecule. Such an approach has been used to image
biofilm formation by attaching a fluorochrome dye to
the acceptor molecule (Koo et al., 2010). It is cer-
tainly conceivable that a similar approach could be
used to deliver therapeutic agents and thereby
ensure that they are retained in the mouth at a clini-
cally relevant site. The advent of nanoparticles has
opened numerous possibilities of delivering therapeu-
tic agents. Several distinct moieties can be attached
to the surfaces of the nanoparticle and released, for
example at low pH values. The surface properties
can be formulated to ensure that they are retained on
the teeth or mucous membranes (Horev et al., 2015).
Clearly additional avenues could be followed to
explore methods to prevent dental caries. These
include, but are certainly not limited to, killing of
specific organisms, replacement of virulent organisms
by non-virulent mutants, use of probiotics, and the
release of therapeutic agents from restorative materi-
als. The concept of implanting microorganisms into
the mouth is likely to be greeted with some resis-
tance. It is recognized that microorganisms from the
mouth may be associated with bacterial endocarditis.
The risk, albeit small, of an implanted organism colo-
nizing a heart valve would hardly be acceptable. Cer-
tainly the use of highly selective homing agents, such
as lectins or antibodies, to deliver therapeutic sub-
stances to specific sites in the mouth, including bac-
teria, continues to be explored. Using restorative
materials to deliver therapeutic agents is essentially a
refinement of the slow-release concept. A difficulty
may arise during long-term use; as the therapeutic
agent is released from the restorative material it will
be replaced by water (saliva), usually to the detriment
of the integrity of the restorative material. Neverthe-
less, the concept could be used in special clinical sit-
uations, e.g. patients with rampant carious lesions.
Overall, dental caries in most instances is an ‘over-
the-counter’ disease and the use of prescription ther-
apeutic agents with added cost is unlikely to be
widely accepted. If, on the other hand, we can con-
vince patients and the dental profession that the
insertion of a restoration, however necessary, is not
treating the disease, dental caries, perhaps more
sophisticated methods of prevention might be accept-
able. As already alluded to, enhanced methods of
delivery remain a considerable challenge. By
232
W.H. Bowen
increasing our understanding of the molecular mecha-
nisms involved in pellicle formation, in particular the
contribution of bacterial products, of biofilm initiation
and maturation, interactions among bacteria in plaque
and the forces that maintain the structure of the
matrix, additional avenues for effective delivery of
agents to prevent and treat dental caries will become
available. It is apparent that there is more to dental
caries than cavitation!
REFERENCES
Appleton, J.L.T. (1944) Dental Caries. Bacterial Infection,
with Special Reference to Dental Practice. London:
Henry Kempton, pp. 332.
Bowen, W.H. (2013a) The Stephan Curve revisited. Odon-
tology 101: 2–8.
Bowen, W.H. (2013b) Rodent model in caries research.
Odontology 101: 9–14.
Bowen, W.H. and Koo, H. (2011) Biology of Streptococcus
mutans-derived glucosyltransferases: role in extracellu-
lar matrix formation of cariogenic biofilms. Caries Res
45: 69–86.
Bowen, W.H., Amsbaugh, S.M., Monell-Torrens, S. et al.
(1980) A method to assess cariogenic potential of food-
stuffs. J Am Dent Assoc 100: 677–681.
Falsetta, M.L., Klein, M.I., Colonne, P.M. et al. (2014)
Symbiotic relationship between Streptococcus mutans
and Candida albicans synergizes virulence of plaque
biofilms in vivo. Infect Immun 82: 1968–1981.
Flemming, H.C. and Wingender, J. (2010) The biofilm
matrix. Nat Rev Microbiol 9: 623–633.
Horev, B., Klein, M.I., Hwang, G. et al. (2015) pH-acti-
vated nanoparticles for controlled topical delivery of far-
nesol to disrupt oral biofilm virulence. ACS Nano 9:
2390–2404.
Ko€hler, B. and Andre en, I. (2012) Mutans streptococci and
caries prevalence in children after early maternal caries
prevention: a follow-up at 19 years of age. Caries Res
46: 474–480.
Koo, H., Xiao, J., Klein, M.I. and Jeon, J.G. (2010) Ex-
opolysaccharides produced by Streptococcus mutans
glucosyltransferases modulate the establishment of
microcolonies within multispecies biofilms. J Bacteriol
192: 3024–3032.
Liao, S., Klein, M.I., Heim, K.P. et al. (2014) Streptococ-
cus mutans extracellular DNA is upregulated during
growth in biofilms, actively released via membrane vesi-
cles, and influenced by components of the protein
secretion machinery. J Bacteriol 196: 2355–2366.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
W.H. Bowen
Meyerowitz, C. and Watson, G.E. (1998) The efficacy of
an intraoral fluoride-releasing system in irradiated head
and neck cancer patients: a preliminary study. J Am
Dent Assoc 129: 1252–1259.
Robinson, C. (2011) Mass transfer of therapeutics through
natural human plaque biofilms: a model for therapeutic
delivery to pathological bacterial biofilms. Arch Oral Biol
56: 829–836.
Svensa €ter, G., Borgstro
€m, M., Bowden, G.H. and Ed-
wardsson, S. (2003) The acid-tolerant microbiota associ-
ated with plaque from initial caries and healthy tooth
surfaces. Caries Res 37: 395–403.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Molecular Oral Microbiology 31 (2016) 228–233
Dental caries – a perspective
Tanzer, J.M., Livingston, J. and Thompson, A.M. (2001)
The microbiology of primary dental caries in humans.
J Dent Educ 65: 1028–1037.
Vacca-Smith, A.M. and Bowen, W.H. (1998) Binding prop-
erties of streptococcal glucosyltransferases for hydrox-
yapatite, saliva-coated hydroxyapatite, and bacterial
surfaces. Arch Oral Biol 43: 103–110.
Vacca-Smith, A.M., Venkitaraman, A.R., Quivey, R.G. Jr
and Bowen, W.H. (1996) Interactions of streptococcal
glucosyltransferases with alpha-amylase and starch on
the surface of saliva-coated hydroxyapatite. Arch Oral
Biol 41: 291–298.
233