0003-9969,92 55.00 + 0.00 Copyright p 1992 Pcrgamon Press Ltd
Archs oral Biol. Vol. 37, No. I I, PP. 913-922, 199: Printed in Great Britain. All rights reserved
pH RESPONSES TO SUCROSE AND THE FORMATION pH GRADIENTS IN THICK ‘ARTIFICIAL MOUTH’ MICROCOSM PLAQUES C.
OF
H. SISSOSS, T. W. CUTRESS, G. FAULDS* and L. WOXG
HRC Dental Research Unit, Wellington School of Medicine, Otago University, P.O. Box 27-007, Wellington, New Zealand (Receiced 11 February 1992; accepted 27 Ma? 1992) Summary-Artificial microcosm plaques were grown in a five-plaque culture system for up to 6 weeks, reaching a maximum depth of several mm. Procedures for long-term pH measurement with glass electrodes were established; they showed that the application of 5 or 10% sucrose for 6 min with a slow continuous flow of a basal medium containing mucin (BMM) generated the pH changes characteristic of in rho Stephan curves. These pH responses were reproducible between plaques. Plaque mass and thickness were critical variables. Successive, sucrose-induced pH curves in plaques up to 4 mm thickness showed minor reductions only in the amplitude and rates of pH change. In plaques over 4 mm thick there was a pronounced reduction in pH response to successive sucrose applications, indicating increased diffusion limitations-a result of plaque growth to seal in the freshly-inserted pH electrode. In plaques of 6 mm maximum thickness, 10% sucrose induced a decrease to below pH 5.5 lasting 24 h, compared to the pH response in 2 mm thick plaque, which returned to the resting pH in 2 h. Differences in pH of up to 0.9 units were identified in thick plaques between inner and outer layers. The BMM flow rate was a critical determinant of the amplitude of the pH response to sucrose and subsequent return to resting pH. These results confirm, for microcosm plaque, the importance of clearance dynamics and diffusion-limited gradients in regulating plaque pH.
Key words: pH, microcosm plaque, sucrose. Stephan curve, diffusion, plaque pH gradients,
et
al., 1989: Macpherson and Dawes, 1991b; Macpherson, Chen and Dawes, 1991). It was The pH is a dominant factor in the pathology and concluded that critical variables of the film affecting ecology of dental plaque. Acidic pH fluctuations pH responses include flow rate and turnover, which in plaque induced by dietary carbohydrate have a control the supply and removal of substrates, metabtypical profile, the Stephan curve (Stephan, 1944; olites and buffers (Lagerlof, Dawes and Dawes, 1984; Kleinberg and Jenkins, 1964). Factors affecting the Dawes, 1989). and plaque thickness (Dawes and magnitude and shape of the curve include the rate Dibdin, 1986; Dawes, 1989, Dibdin 1990a, b). One approach to modelling complex microbial ecoand duration of: acidogenesis (which in turn is related to the supply, concentration and removal of carbosystems such as dental plaque is to study laboratory hydrate); alkali generation; CO, loss, removal of submicrocosms--culture systems where the mixed natural strates, metabolites and buffers into the surrounding flora is evolved in vitro under controlled environmental fluid environment, and from the mouth (Kleinberg and nutrient conditions (W5mpenny, 1988). Plaque et al., 1982; Schachtele and Jensen, 1982; Dawes, microcosms cultured in vitro from the mixed oral flora 1989; Dibdin, 1990a, b). Interactions with plaque, have advantages in investigating plaque behaviour enamel, saliva and gingival crevicular fluid buffers (Sissons, Cutress and Pearce, 1985; Sissons, Hancock are important (Dibdin, 1990b). and Cutress, 1988; Sissons et al., 1991). In order Substantial progress in quantifying the importance adequately to control and replicate experiments in of the processes involved has followed the discovery plaque metabolism and ecology it is necessary to that exchanges between supragingival plaque and the monitor, and ultimately, to manipulate plaque pH. oral environment are mediated by a thin film of saliva We have developed a five-plaque ‘artificial mouth’ (Collins and Dawes, 1987). The flow rate of this saliva microcosm for such studies in which plaques can be film and its exchange with bulk saliva vary at different grown in controlled environmental conditions, includsites (Dawes et al., 1989). Stephan pH curves resulting ing those of fluid and nutrient composition and supply. from diffusion of carbohydrate into plaque from this This provides a model system for examining the saliva film, its metabolism and end-product removal e,ffect on pH of many variables that cannot be readily have been studied in simple, in vitro models (Dawes studied in viva (Sissons et al., 1991). A continuous supply of a mucin-containing, peptone-vitamin nutrient solution, and the periodic addition of su*Present address: Department of iviolecular Pharmacology, crose simulates the in vivo dynamics of carbohydrate supply and removal, such as with the ingestion of Middlesex and Hospitals Medical School, Windeyer Building, Cleveland Street, London WIP 6BD, U.K. sucrose drinks. Because microcosm plaques model Iir(TRODL’CTION
913
C. H. Stsso~s et al.
91-t
the composition of natural plaque and hence its behaviour more closely than do simple, defined systems based on chemical exchanges or single organisms. they further test the involvement and significance of various processes in pH control. Our overall objective is to understand and control pH in plaque microcosms. Specific aims now were first to establish procedures for long-term, continuous, reproducible pH measurement in ‘artificial mouth’ microcosm plaques, and second to examine pH changes generated by transient exposure to sucrose and the effects on them of plaque thickness and of variation in the flow rate of a continuously supplied, peptone-mucin nutrient.
hI.-\TERIALS
AND METHODS
Plaque culture apparatus and growth procedures
Microcosm plaques were grown in a five-plaque artificial mouth as described by Sissons et al. (1991). It consists of: a horizontal. 110 mm dia cylinder with a detachable end-plate: five plaque-growth stations 80 mm apart; gas and thermometer ports on the top; a port in the bottom at one end leading to a 20-I waste reservoir and gas scrubbers. A later version of the culture chamber with endplates at both ends and larger side-ports was also used. Each plaque-growth statton included in a planar section of the growth chamber: opposed horizontal ports, one a B34 cone through which the intact plaque holder could be inserted; a vertical B24 port for the fluid delivery head; and angled ports on opposite sides for inoculation (B24, 60’ above horizontal) and sampling (B40, 45’ above horizontal). Reference and pH electrodes, respectively, were also installed through these angled ports. A revised fluiddelivery head was also used, which included, in addition to the fluid lines, an injection port and a drop-positioning rod to give more precise application of fluid to the plaque (Plate Fig. 1). Plaque growth was initiated by inoculation with whole mixed saliva on days 0, 3 and 5 of growth. A basal medium containing 0.25% mucin (BMM), based on the medium of Glenister et al. (1988) was supplied continuously at 3.3-3.6 ml/h to each plaque under standard conditions. It contained: 0.5% trypticase peptone, 1.0% proteose peptone. 0.5% yeast extract (all from Difco Laboratories. Detroit, MI. U.S.A.), 0.25% KCl, 5 mg/l haemin. 1 mg/l menadione, 0.25% partially purified pig gastric mucin (type III. Sigma Chemical Co., St Louis, MO, U.S.A.). In some experiments (see figure legends), the BMM was modified by supplementation with arginine (1 mmol~l) and urea (0.5, 1 or 20 mmol/l).
Every 6 or 8 h, unless otherwise stipulated, 1.5 ml of 5% w/v sucrose were supplied to each plaque over 6 min in addition to the BMM. Plaques were grown for up to 6 weeks. Experiments were made to establish the reproducible measurement of sucrose-induced pH curves, and to examine the effects of plaque mass, thickness and BMM flow rate on them. Intraplaque pH gradients during responses to sucrose were examined. Plaque support configuration
Plaques were grown in different physical configurations. The plaque supports were usually plastic, 25 mm dia. Lux Thermanox (TM) coverslips, which, for initial experiments. were attached to glass platforms as described by Sissons et al. (1991). A revised holder is now standard, where the coverslips are supported on glass rings with retaining and drainage posts [Plate Fig. 2(A)]. These arrangements allowed unconstrained plaque growth over the coverslip until the excess fell periodically to the base of the culture chamber. In addition, two plaque-support systems giving more constrained growth were developed. One was primarily for holding enamel blocks and consisted of a 16 mm square, 5 mm deep glass basket supporting either a trimmed Lux coverslip or 4 mm deep block of polyester resin (Epiglass 552P polyester encapsulated resin, Epiglass NZ Ltd, Auckland. New Zealand), usually containing four, embedded. polished, 3 x 4 mm, bovine enamel blocks [Plate Fig. 2(B)]. To examine the effect of plaque thickness. a 2.5 mm thick polyester spacer (no enamel) was placed under the trimmed coverslip to give a plaque 2.5 mm deep, as compared with a 5 mm thick plaque on a coverslip only. An approximately constant thickness was maintained by passing a Teflon blade ‘sweeper’ [Plate Fig. 2(C-iii)] across the holder. The second was a prototype constant-thickness holder [Plate Fig. 2(C-i)] consisting of a Teflon block, 36.5 mm long, 31.5 mm wide. 12.5 mm thick, threaded on to a plastic-coated steel rod. The block had a 25 mm dia hole into which tightly fitted a 10 mm thick Teflon plug with a 5 mm hole through its centre on which rested a Lux coverslip. The depth above the coverslip was set with a steel spacer [Plate Fig. 2(C-ii)] as described by Peters and Wimpenny (1988). To give drainage, four equidistant, vertical, 5 mm dia holes were drilled with their centres offset 1 mm from the rim of the central recess containing the plaque. The plaque surface was hand swept daily in the culture chamber with a 34 mm wide, 1 mm thick Teflon blade in a holder fitted with a guide to stop lateral movement [Plate Fig. 2(C-iii)], and plaque present in the drainage holes was removed.
Plate 1 Fig. 1. Fluid delivery head for plaque-growth station. This fitting is based on a B24/29 cone and is modified from the head used before (Sissons et al., 1991) as follows. The top was expanded into an approx. 30 mm dia chamber with angled tubes to take lines for sucrose and experimental treatments. an angled port closed with a Schott LC 14 cap and silicone septum for direct injection of fluids, and the main (BMM) feedline shortened to stop inside the chamber to form an infection lock. Below the B31 29 cone, the glass was shaped into a cone leading to a 4 mm dia hole (A) which was directly over a 5 mm dia bulge in the glass, 2 mm dia. drop-positioning rod. Two further holes (B and C) were positioned to allow fluid overflow and give gas-phase access if hole (A) was blocked. The end of the positioning rod was at a height of 15 mm above the Lux coverslip and positioned drops of fluid centrally over the plaque.
MicrocosmplaquepH
Plate I
C. H. Sisso?;s ei ul
Plate 2
917
Microcosm plaque pH Table I. Effect of 5 and 10% sucrose on pH change variables in plaque Variable (SD*) Initial pH pH minimum APH Minutes to pH minimum Area of curve (pH x h) Area of curve below pH 5.5 (pH x h)+
5% Sucrose 6.33 (0.13) 5.75 (0.07) -0.57 (0.07) 35.9 (8.8) 1.37 (0.20) 0
10% Sucrose
Ratio 5 IO(%)
6.32 (0.05) 5.26 (0.09) - 1.OS(0.05) 41.8 (4.5) 2.17 (0.30) 0.13 (0.06)
0.54 0.86 0.62 0
*Six replicate curves induced by 5% w Y sucrose were followed by six induced by 10% w L sucrose in :he same nlaoue, with a BX4MRow rate of 3.6mlih. +Taken as a measure of the ‘critical pH’.
pH monitoring
Reproducibility
A micro-oesophageal glass pH electrode (1.4 mm dia. Model SA3. World Precision Instruments, New Haven, CT, U.S.A.) mounted in stainless-steel tubing, and a micro-reference electrode (Model 401, Diamond General Co.. Ann Arbor. MI, U.S.A.) were installed in the culture chamber and plaque, at angles of approx. 45 and 60’ from the horizontal. They were connected through a pH meter (Model 1300. Diamond General Co.) and custom-built signal conditioner to a chart recorder (Model 8100. Philips, Eindhoven. The Netherlands). The reference electrode was positioned near the edge of the plaque and the pH electrode well into the plaque. or in various positions (see figure legends). The electrodes were calibrated periodically ivith pH 7 and 4 buffers.
Sucrose applied rapidly over 6 min (I5 ml h per plaque) to plaques grown with BMM-urea ( 1 mmol,l) arginine (I mmol/l) induced a decrease followed by a slow return to the resting pH, characteristic of Stephan curves (Text Fig. 3). In the absence of added substrates, the pH reached a stable, steady-state, resting pH that typically varied by only 0.02 pH unit over, for example, 9 h. Larger pH fluctuations sometimes occurred, probably associated with changes in plaque mass as excess plaque grew and fell from the growth station. After the insertion of the pH electrode. successive, sucrose-induced pH changes were reproducible, obtained in all but very thick plaques [Text FIN. 3(A)]. Small changes in shape of successive curves did occur in thin plaques; these were pronounced in thicker plaques [Text Fig. 3(B)]. In the 2.5 mm plaque, growth in volume was minimal, but did fill in any gap
AnalJ,tical
techniques
Plaque depth was measured lvith a periodontal probe calibrated in mm. Changes in pH were analysed using the minimum and resting pH and the maximum pH change (ApH in pH units) (Sissons and Cutress, 1987). The area of the sucrose-induced pH curve, or the area below pH 5.5 was calculated in units of pH x time (h) when appropriate.’
RESULTS Electrode
performance
After initial conditioning of a pH electrode in plaque, the slope between pH 7 and 4 remained 3 f 0.1 pH units for over 2 weeks, indicating stability of the pH electrode. The E’ (pH 7) value drifted more, a result of changes in the reference electrode. In a well-stabilized system examined after 18 days of continuous measurement, the E’ had drifted 0.08 pH unit, and the response between pH 7 and 4 had changed 0.05 pH unit. Different reference electrodes varied in stability. response and longevity.
of sucrose-induced
pH changes
caused by inserting the pH electrode. This reduced the size and rate of pH decrease and the rate of rise
by reducing access of the applied sucrose. Reproducible pH changes were obtained by inserting electrodes into plaque at daily intervals o\er 5 days [Text Figs 3(C) and (D)]. indicating that disturbing normal-thickness plaques by such insertion did not invalidate pH measurement. This allows for external electrode recalibration. In five replicate plaques grown for 2 weeks in BMM-urea (1 mmol,I)-arginine (1 mmol 1) with 5% sucrose f&hourly, the resting pH was 6.39 (SD 0.08) and the minimum pH induced by 5% sucrose was 5.83 (SD 0.10) (data not shown). Comparison
of pH responses
to 5 and 10”0 sucrose
The pH response of plaque to 10% sucrose was almost twice that of 5% sucrose (Table I). Effeect of plaque mass and thickness on sucrose -induced pH curves
The shape of sucrose-induced pH curves varied with plaque age and hence mass, and with different
Plate 2 Fig. 2. Plaque-support systems. (A) Revised, unrestrained growth, plaque-support system. The Lux coverslip rested on a 2 mm thick. 25 mm dia glass ring mounted on a glass rod with equidistant 2 mm dia glass rods raised approx. 2 mm above the ring surface to locate the coverslip and break the fluid meniscus of the plaque, and corresponding posts, reaching about 4mm below the ring, to assist in the formation of reproducible drops and hence in plaque drainage. (B) Glass-basket, plaque-support system containing four bovine enamel blocks. (C) Prototoype, constant-thickness plaque holder (i) containing a Lux (TM) coverslip, with final plaque thickness being set by the 2.0mm spacer (ii). The Teflon-blade sweeper (iii) is used to remove excess plaque.
918
C. H. SISSONSet al.
5 I
7
I
I
I
I
I
I
I
I
6) s
I
I
I
I
I
3
4
5
6
(B)
:
6 I Q 5
:q 1
1
4
0
I 12
I
I
I
I
I
I
3
4
5
6
7
12
0
Hours
1. ; !
7 1 Qa 2 m $
2.0
i? 5 0.4
1.5
1.0
& > 2
0.2
2
0.5
z
0.1
I a
0.0
: 2
0.0
12345
Day
7
Hours
Ia
0.3
5
&J, 1
12345
Day
Fig. 3. Reproducibility of sucrose-induced pH responses. (A) Four successive pH curves induced by 5% w/v sucrose (1.5 ml over 6 min, indicated by the bar labelled S) after initial pH electrode insertion into normal-thickness plaque (4mm maximum) grown in a basket holder with BMMurea (1 mmol/l)-arginine (I mmol 1) at 3.3 ml,‘h per plaque for 4 weeks with 6-hourly 5% w, v sucrose. The substratum was polyester resin containing four bovine enamel blocks. (B) Successive, sucrose-induced pH curves after pH electrode insertion into thick plaques. and effect of plaque mass. Plaques grown for 25 days with BMM-urea (1 mmol/l)arginine (I mmol/l) at 3.3 ml/h per plaque and 6-hourly 10% w/v sucrose were mixed and deposited to 2.5 mm thickness on a trimmed Lux coverslip in a basket holder. The plaque mixture was grown for 4 days with I-hourly 10% w/v sucrose, the surface trimmed and the pH electrode inserted near one corner. pH changes were induced by d-hourly 10% w/v sucrose, 1.5ml over 6 min as shown by the bar. After the first sucrose-induced pH curve (2 x PLQ, . . .) approx. 50% w/v of the plaque was removed from the area distant from the pH electrode, leaving the plaque around it undisturbed and the same thickness. The following sucroseinduced pH curve (30min after plaque removal) is shown (1, -), and curves 24 h (2, ---) and 48 h (3, -) later. During this period the plaque mass was not visibly increased. (C) The pH decrease (ApH) induced by 5% w/v sucrose (1.5ml in 6 min) over 5 days, measured by daily reinsertion of electrodes into the same plaque. The mean pH decrease was 1.31 (SD 0.11) pH units. The plaque had been grown for 12 davs in BMM-urea (1 mmol’l)-arginine (1 mmol/l) at 3.3 ml/h per plaque on enamel in polyester resin (as in A). (D) Areas of sucrose-induced pH curves during the 4 h after sucrose addition for the pH decreases shown in (C). The mean area was 0.296 (SD 0.033) pH x h.
0
10
r, 20 30 Hours
40
!I
Fig. 4. Effect of plaque thickness on sucrose-induced pH changes. (A) Sucrose-induced pH changes in plaque at different thickness (and growth stage) grown in BMM-urea (0.5 mmol/l)-arginine (I mmol/l), 3.6 ml/h per plaque. Five % sucrose was applied, 1.5 ml over 6 min, as shown by the bar labelled S. Curve I (-) shows the pH changes in a 2-weeks, approx. 4 mm plaque. Most plaque was removed except for a thin layer and curve II c ,) shows the pH curve after 2 days of regrowth to approx. 1.5 mm thick. (B) Representative pH response to 5% w/v sucrose (1.5 ml over 6 min at the time shown by the bars) in a plaque grown for 27 days in BMM (3.6ml/h per plaque) whilst constrained to constant thickness (2.5 mm) by being swept daily (-). Also shown is a plaque grown for the same time but with unrestrained growth to approx. 4 mm maximum thickness (---), (C) pH changes induced by 10% w/v sucrose in a very thick (6 mm maximum) plaque. The plaque was 33 days old, grown on a 25 mm dia Lux coverslip mounted on a glass-platform holder in BMM for 2 weeks and then in BMM-urea (20 mmol/l) at 3.6 ml/h for the remainder of the period. The pH electrode was positioned centrally in the plaque 2 weeks previously. The arrow indicates time of 10% w/v sucrose addition (1.5 ml in 6 min).
types of plaque holder. Plaques grown unrestrained on coverslips were convex and generally increased in size to a maximum that depended, for example, on the impact of the drops of fluid supplied and on shelter by electrodes. Here, sucrose-induced pH curves ranged from a rapid decrease and return to the resting pH in freshly grown plaques, to a slower, smaller decrease that required over 4 h to return to resting pH in thicker and older plaques [Text Fig. 4(A)].
919
Microcosm plaque pH
I
n
1: I
;
1
I
I
0
12
I
I
I
I
I
I
3
4
5
6
7
Ia
I
I
I
I
I
I
I
I
I
_
_p)
6 qy-
Hours 5 (6)
3tV1 0
12
6.0
I 3
I 4
I
I
I
5
6
7
6
Hours
r-l % 0
0
-7&h!&-+
BMM flow rate (ml/h/plaque)
Fig. 5. The effect of BMM flow rate on sucrose-induced pH changes. pH changes induced by 10% w/v sucrose (1.5 ml over 6 min. shown by the bar labehed S) in a plaque (3-4 mm maximum thickness) previously grown on BMM for 9 days on a 25 mm coverslip (glass-platform holder) were examined over 5 days at BMM flow rates of 0, 1.8, 3.6, 7.2 and 18 ml/h per plaque. These different flow rates diluted the sucrose to 10.0,8.9,8.1,6.8 and 4.6%. Replicate curves were examined at intermediate flow rates. (A) Representative pH changes at BMM flow rates of 0, 1.8, 7.2 (-, as labelled) and 18 ml/h ( .). (B) The sucrose-induced minimum pH at different flow rates.
Removal of half the plaque mass increased the amplitude and rates of pH change [Text Fig. 3(B)]. When the growth of plaques to a constant size and thickness was examined in a constant-thickness holder where plaque above 2.5 mm was removed daily [Text Fig. 4(B)], the pH decrease in plaque of constant 2.5 mm thickness was greater than in similarly aged plaques with unrestrained growth (4-5 mm, maximum depth). In another experiment, pre-grown plaque was deposited to give 2.5 and 5mm thicknesses in the baskets designed for experiments on enamel [Plate Fig. 2(B)]; 8 h after sucrose exposure, the 5 mm plaque was still pH 4.6 whereas the 2.5 mm plaque had returned from a pH 5.0 minimum to its resting pH of 6.4. Very thick plaques (e.g. 7 mm maximum depth) formed on the 25 mm dia Lux coverslips after several weeks of growth could give long-lasting pH changes. For example, in a 6mm plaque, the pH changes induced by 10% sucrose [Text Fig. 4(C)] had a minimum lasting 15 h below pH 5, and took 2 days to return to the resting pH, despite the presence of 20mmol/l urea in the BMM (which had raised the resting pH to 7).
Fig. 6. pH gradients formed during sucrose metabolism. Plaque was grown in BMM-urea (1 mmol/l)-arginine (1 mmol/l) at 3.3 ml/h per plaque on 25 mm dia Lux coverslips. It reached approx. 7 mm depth in 16 days. The pH electrode was located in the top surface layers (T) of the plaque or at the base (B). The bar shows 5% w;‘v sucrose addition (1.5 ml over 6 min). (A) Five % sucrose-induced pH changes at the base and surface of the plaque. For the curve labelled B (-), the pH electrode was inserted deep to the plaque base 12 h before the addition of sucrose. The curve labelled T (-) shows the second sucrose-induced pH curve in which the pH electrode was located in the plaque surface layers. (B) The pH response measured with the pH electrode repositioned between the plaque surface and the plaque base before and during the pH fluctuation. These sucrose-induced pH responses were flanked by the pH response curves shown in (A), curve B which is shown as a dashed line. The pH electrode was positioned initially at the base of the plaque at point l(B). The pH electrode was raised to the top surface layers of the plaque, point 2(T). The surface pH was recorded and then sucrose added. After the rapid surface pH fall during sucrose application [to point 3(T)], the pH electrode was inserted to the base of the plaque. The higher base pH was measured at point -I(B’). The pH electrode was then repositioned in the plaque surface, point 5(T’), where it measured the rise portion of the Stephan curve in the plaque surface layer.
Effect of BMMflow rate on sucrose-inducedpH
curves
Limiting the flow rate of BMM in normal (2-4 mm maximum thickness) plaques had a substantial effect on the Stephan curve (Text Fig. 5). If the supply of BMM was stopped at the time of sucrose addition, the pH fell and remained low [Text Fig. 5(A)]. Increasing the BMM flow rate reduced the amplitude of the pH change until further increases in flow rate made little difference. The minimum pH reached in the Stephan curve proved to be the most replicable measure of the pH changes in these experiments. It increased with increasing BMM flow rate to a maximum value [Text Fig. 5(B)]. The initial pH was largely a function of the shape and duration of the previous sucrose-induced pH curve and varied significantly during these experiments on pump-rate variation. Differences in initial pH invalidate direct comparisons with measures such as the magnitude of the pH decrease and the area of the pH curve. Plaque pH gradients
The growth of plaques several mm thick provided an opportunity to examine intraplaque pH gradients.
920
C. H. S~sso~set al.
AS fully developed plaques on 25 mm dia coverslips neared the resting pH. there was a small but significant 0.1-0.2 pH unit gradient from the edge to the centre of the plaque at the base. However, in a thick (7 mm maximum) plaque in response to metabolism of 5% sucrose, substantial pH gradients were formed between the top surface of the plaque centre and its base. After sucrose application, the pH response on the plaque surface was a typical Stephan curve compared with virtually no pH change in deep plaque against the coverslip [Text Fig. 6(A)]. This pH gradient was also demonstrable by moving the pH electrode between the plaque surface and base during a Stephan curve [Text Fig. 6(B)]. The electrode in the plaque surface during sucrose application showed a sucrose-induced decrease at the surface of 0.95 pH unit. Inserting the electrode to the base of the plaque indicated that there was a sucrose-induced decrease of 0.06 pH unit only. A comparison with Text Fig. 6(A) shows that moving the electrode caused some mixing of surface and base plaque pH conditions. DISCUSSION
Long-term
plaque pH measurement
No ideal procedure for the long-term measurement of plaque pH in situ has yet been devised. Calibration,
drift. and the response of the measurement system to the plaque environment are issues that need to be addressed in in vitro (Russell and Coulter, 1975; Hudson, Donoghue and Perrons, 1986) and intraoral studies (Kleinberg et al., 1982; Schachtele and Jensen. 1982; Igarashi, Lee and Schachtele, 1989, 1990) of plaque pH. In cico experiments are limited to a few days and give little possibility for controlling plaque formation or environment. In clitro experiments can be extended to many weeks, but calibration problems are increased. The drift in the response of our electrode system, though small, necessitated occasional recalibration. Periodic removal of the electrodes and calibration checking has the advantage of confirming the reliability of the system overall. Limitations include disruption of plaque structure and difficulty in accurately positioning the pH electrode. In thick plaques disturbed by inserting electrodes, the size and rate of pH decrease with successive sucroseinduced pH changes indicates increased diffusion barriers as a result of plaque growth. Nevertheless, the pH cumes measured were relatively reproducible, especially considering that plaque morphology and mass were usually undefined. .\ficrocosm plaques natural plaque
as a model for pH regulation
in
Studies of carbohydrate-induced, plaque pH changes involving computer and simple, in vitro, monobacterial models have indicated that limitations on diffusion set up interacting intraplaque gradients of pH, acid- and alkali-generating substrates, acids, ammonia and mobile buffers (Dawes and Dibdin, 1986: Dawes, 1989, Dibdin, 1990a, b; Macpherson and Dawes, 1991a, b; Macpherson et al., 1991). Plaque pH is determined by the supply, metabolism and clearance dynamics of these compounds, and their
interaction with bacterial cell buffers and enamel. Plaque thickness, and the rate of supply of oral fluids, generally saliva, are major variables affecting the outcome. Microcosm plaques are distinct from defined bacterial systems. They are similar in appearance, structure. composition and behaviour to natural plaques (Sissons et al., 1991); our results confirm this for pH behaviour in response to sucrose. Variables critical to pH control can be studied in microcosm plaques with unrestrained growth, and potentially studied in plaques of more defined size and shape. A sucrose rinse lasting a few minutes, followed by sucrose and metabolite clearance, simulates the processes that arise in riro during the intake of sucrose drinks (Lagerlof et al., 1984; Sreebny, Chatterjee and Kleinberg, 1985: Lindfors and Lagerlof, 1988). This rinse is commonly used in modelling cariogenic challenges (Dawes, 1989; Igarashi et al., 1989; 1990). The pH changes typical of Stephan curves in the microcosm plaques were usually of longer duration than the curves described for smooth-surface enamel plaque (e.g. Kleinberg et al., 1982; Schachtele and Jensen, 1982; Igarashi et al., 1989. 1990). presumably because of the size of the microcosm plaques. The increased magnitude of the pH response to 10% sucrose compared to 5% is similar to that predicted in computer models (Dibdin, 1990a. b). reported in model systems (Lagerlof et al., 1984), and for in riro plaques (Lindfors and Lagerlof, 1988). Replicate plaques Fave a highly reproducible resting pH that was raised If urea was present in the BMM (Sissons et al., 1991). as shown in Text Fig. 4(C). Replication of sucrose-induced pH curves in replicate plaques was good; any variation can be explained by a sensitive dependence on the stage of plaque growth and its size. Minor variations in drainage seemed to affect the pattern of plaque growth and would potentially affect intraplaque gradients, clearance of substrates, metabolites, and hence pH. One reason that the minimum pH was the most repeatable measure of the sucrose-induced pH curve is that the pH scale is a logarithmic transform of hydronium ion concentration, which would tend to minimize experimental variation at low pH. This may be countered to a degree by the high buffering capacity of plaque (Stralfors, 1948; Sissons and Cutress, 1987, Shellis and Dibdin, 1988; Dibdin, 1990b; Macpherson and Dawes, 1991b), which would tend to make the pH response linear. Plaque sire, thickness and Stephan curces
The potentially profound effect of plaque thickness on the Stephan pH response has been indicated by computer modelling (Dawes and Dibdin, 1986; Dawes, 1989; Dibdin, 1990a, b). The thickness of natural dental plaque varies with location. On smooth enamel surfaces it ranges up to 200pm (Main et al., 1984). However, in fissures and approximal sites, plaque is potentially l-2 mm thick as measured from the surface exposed to saliva. These are sites of high caries susceptibility. Intra-oral models of fissure plaque have ranged up to 2 mm depth (Igarashi et al., 1989). There seems to be little published information on the size and thickness of approximal plaques but where oral hygiene is poor, and tooth or restoration configuration appropriate, they potentially grow to
921
Microcosm plaque pH
virtually the full width of molars. The inner regions of such plaques are inaccessible to saliva exchanges, and Stephan curves in them can remain at a low pH for long periods (Schachtele and Jensen, 1982; Igarashi et al., 1989, 1990). Microcosm plaques grow up to several mm thick. Those of 6mm thickness, with sucrose-induced pH curves below pH 5 for over 12 h, probably reflect a more extreme situation than occurs in vivo. They provide an opportunity to study pH behaviour at the limits of diffusion restrictions and mode! this aspect of approximal and fissure plaques. The size and shape of plaques in the mouth are constrained by movements of the tongue, lip and cheeks, by mastication and by oral hygiene procedures. It would be desirable to simulate these processes in vitro to produce plaques of defined morphology. A daily sweep with a Teflon blade over a 25 mm dia, constant-thickness plaque holder defined plaque thickness and resulted in sucrose-induced pH curves of greater amplitude than in free-growing plaques. How often the actively growing plaque surface is removed will affect the selective pressures on plaque growth. Allowing fluid drainage from the plaque proved important in achieving a relatively solid biofilm. It avoided the submersion of plaque in a substantial liquid phase (Peters and Wimpenny, 1988) and increased the clearance of soluble substrates and metabolites including toxins. More in vitro and in riro research is needed before it will be possible to simulate realistically in citro the processes that physically limit plaque in ciro. BMM jlow rate and Stephan curves Applying sucrose at 15 ml/h per plaque ensured that there was only modest dilution by the BMM at normal flow rates, resulting in a limited effect on the rate of pH decrease during sucrose application. Subsequently the BMM flow rate set the clearance characteristics of the system and proved to be of critical importance in modifying the shape of the Stephan curves. With no flow at al!, the pH decrease was large and the pH did not rise from the minimum. At higher flow rates the pH decrease was very small. Under our standard growth conditions, the BMM flow rate of 3.3-3.6 ml/h per plaque limited the clearance dynamics. The great variation in total pH decrease and curve shape with flow rate (Text Fig. 4) confirms, for microcosm plaques, the dependence of plaque pH behaviour on clearance dynamics into the oral fluid, as demonstrated in simple chemical (Dawes et al., 1989), computer (Dawes and Dibdin, 1986; Dibdin 1990a, b) and monobacteria! mode! systems (Macpherson and Dawes, 199!a, b; Macpherson et al., 1991). Intraplaque
pH gradients
Dawes and Dibdin (1986) predicted from computer modelling based on diffusion theory that restricted saliva supply dynamics to plaques results in diffusionlimited, intraplaque gradients that regulate the pH. This now has been eIegant!y demonstrated in flowcontrolled, thin liquid-film models based on biofilms of bacteria suspended in agarose (Macpherson and Dawes, 199la, b; Macpherson et al., 1991). There
are also steep oxygen gradients in plaque. such that anaerobiosis is likely within a few hundred pm of a plaque surface exposed to atmospheric oxygen concentrations (Globerman and Kleinberg, 1979). To the best of our knowledge the 0.9 pH unit gradient resulting from a substantial 5% sucroseinduced pH response on the surface of a thick microcosm plaque and the very small pH fluctuation in inner plaque constitute the first reports of large pH differences in a system which approximates to that of natural plaques in composition. This further confirms the existence in dental plaque of the pH gradients that were demonstrated in the studies on Stephan curves in Streptococcus oralis biofilms in agarose (Sfacpherson and Dawes, 199lb; Macpherson et al., 1991). These showed that when the flow of fluid film over plaque is restricted. pH gradients down and along the ‘plaque’ reach over 2 pH units with 10% sucrose. Although in normal-thickness microcosm plaques, 10% sucrose increased the amplitude of the sucrose-induced pH curve (Table 1), in very thick plaques (despite diffusion barriers), the sucrose clearly penetrated sufficiently to lower the pH. However, diffusion limitations reduced the clearance of acid products and thus prolonged the pH minimum and retarded the return to resting pH. A major reason for pH gradients and slow pH responses is the high capacity of plaque cell buffers (Stralfors. 1948; Sissons and Cutress, 1987; Shellis and Dibdin, 1988: Dibdin and Shellis, 1989; Dibdin, 1990a, b; Macpherson and Dawes, 199lb. Macpherson et al., 1991). Whilst diffusion of free molecules may be little restricted in plaque (Dibdin, 1990a, b), they interact with, and have to shift the ionization state of, an approximately 500-fold excess of fixed ionic groups (Sissons and Cutress. 1987). In addition to extracellular stationary and mobile buffers, cell transport processes and intracellular buffering systems may be important in regulating plaque pH. Acknowledgements--We thank E. M. Hancock and P. Smith for technical assistance. J. Teppett, Wellington Public Hospital. designed and constructed the signal conditioner. S. Newcombe and K. McCombe. Victoria Cniversity of Wellington, advised on and did the glass-blowing for the modified culture chamber and fittings. We gratefully acknowledge financial support for this project from L’Oreal Laboratoires de Recherche Appliquee et Dtveloppment, Clichy, France.
REFERESCES
Collins L. M. C. and Dawes C. (1987) The surface area of the adult human mouth and thickness of the salivary film covering the teeth and the oral mucosa. J. dent. Res. 66, 1300-1302. Dawes C. (1989) An analysis of factors influencing diffusion from dental plaque into a moving film of saliva and the implications for-caries. J. dent. kes. 68, 1183-1488. Dawes C. and Dibdin G. H. (1986) A theoretical analvsis of the effects of plaque thickness and initial salivary sucrose concentration on diffusion of sucrose into dental plaque and its conversion to acid during salivary clearance. J. dent. Res. 65. 89-94. Dawes C., Watanabe S., Biglow-Lecombe P. and Dibdin G. H. (1989) Estimation of the velocity of the salivary film at some different locations in the mouth. J. denr. Res. 68, 1479-1482.
922
C. H. SISSO?;S er al.
Dibdin G. H. (1990a) EXect on a cariogenic challenge of saliva/plaque exchange via a thin salivary film studied by mathematical modelhng. Curies Res. 24: 231-238. Dibdin G. H. (1990b) Plaaue fluid and diffusion: studv of \ the cariogenic challenge by computer modelling. J. ient. Res. 69, 1324-1331. Dibdin G. H. and Shellis R. P. (1989) The interpretation of CO, equilibration data to obtain plaque fluid buffer capacities- and comparison with results obtained by titration. J. dent. Res. 68. 1323-1327. Glenister D. A., Salamon Kl E., Smith K. and Beighton D. (1988) Enhanced growth of complex communities of dental plaque bacteria in mucin-limited continuous culture. I
I
Microbial. Ecol. Hlth Dis. 1, 31-38.
Globerman D. Y. and Kleinberg I. (1979) Intra-oral p0, and its relation to bacterial accumulation on the oral tissues. In Sulica and Dental Curies (Eds I. Kleinberg, S. A. Ellison and I. D. Mandel), pp. 275-291. Microbioloy Abstracts, Information Retrieval Inc.. New York. Hudson D. E., Donoghue H. D. and Perrons C. J. (1986) A laboratory microcosm (artificial mouth) for the culture and continuous pH measurement of oral bacteria on surfaces. J. uppl. Bucrerio[. 60, 301-3 10. Igarashi K., Lee I. K. and Schachtele C. F. (1989) Comparison of in uiro human dental plaque pH changes within artificial fissures and at interproximal sites. Curies Res. 23, 417-422. Igarashi K., Lee I. K. and Schachtele C. F. (1990) Effect of dental plaque age and bacterial composition on the pH of artificial fissures in human volunteers. Curies Res. 24, 52-58.
Kleinberg I. and Jenkins G. N. (1961) The pH of dental plaques in the different areas of the mouth before and after meals and their relationship to the pH and rate of flow of resting saliva. Archs oral Biol. 9, 493-516. Kleinberg I., Jenkins G. h’.. Chatterjee R. and Wijeyweera L. (1982) The antimony electrode and its role in the assessment and interpretation of dental plaque pH. J. dew Res. 61, 1139-I 147. Lagerlof F., Dawes R. and Dawes C. (1984) Salivary clearance of sugar and its effects on pH changes by Sueptococcus mirior in an artificial mouth. J. dent. Res. 63, 1266-1270. Lindfors B. and Lagerlof F. (1988) Effect of sucrose concentration in saliva after a sucrose rinse on the hydronium ion concentration in dental plaque. Curies Res. 22, 7-10. Macpherson L. M. D. and Dawes C. (1991a) Urea concentration in minor mucous gland secretions and the effect of salivary film velocity on urea metabolism by Slreptococcus cesfibularis in an artificial plaque. J. periodont. Res. 26, 395-401. Macpherson L. M. D. and Dawes C. (1991b) Effects of salivary film velocity on pH changes in an artificial plaque
containing Srreptococcus oralis, after exposure to sucrose. J. dent. Res. 70, 1230-1234. Macpherson L. M. D., Chen W. Y. and Dawes C. (1991) Effects of salivary bicarbonate content and film velocity on pH changes in an artificial plaque containing Sfreptococcus oralis, after exposure to sucrose. J. dent. res. 70, 1235-1238. Main C., Geddes D. A. M., McNee S. G., Collins W. 1. N., Smith D. C. and Weetman D. A. (1984) Instrumentation for measurement of dental plaque thickness in situ. J. biomed. Eng. 6, 151-154. Peters A. and Wimpenny J. W. T. (1988) A constant-depth laboratory model film fermenter. In CRC Handbook of Laboratory Model Systemsfor Microbial Ecosystems (Ed. Wimpenny J. W. T.), Vol. 1, pp. 175-195. CRC Press, Boca Raton, FL. Russell C. and Coulter W. A. (1975) Continuous monitoring of pH and Eh in bacterial plaque grown on a tooth in an artificial mouth. Appl. Microbial. 29, 141-144. Schachtele C. F. and Jensen M. E. (1982) Comparison of methods for monitoring changes in the pH of dental plaque. J. dent. Res. 61, 1117-l 125. Shellis R. P. and Dibdin G. H. (1988) Analysis of the buffering systems in dental plaque. J. dent. Res. 67, 438-446.
Sissons C. H. and Cutress T. W. (1987) In-&o ureadependent pH-changes by human salivary bacteria and dispersed, artificial-mouth, bacterial plaques. Archs oral f&l. 32, 181-189. Sissons C. H.. Cutress T. W. and Pearce E. I. F. (1985) Kinetics and product stoichiometry of ureolysis by human salivary bacteria and artificial-mouth plaques. Archs oral Biol. 30, 781-790.
Sissons C. H., Hancock E. M. and Cutress T. W. (1988) The source of variation in ureolysis in artificial plaques cultured from human salivary bacteria. Archs oral Biol. 33, 72 l-726.
Sissons C. H., Cutress T. W., Hoffman M. P. and Wakefield J. St J. (1991) A multi-station dental plaque microcosm (artificial mouth) for the study of plaque growth, metabolism, pH and mineralization. J. denl. Res. 70, 14091416. Sreebny L. M., Chatterjee R. and Kleinberg I. (1985) Clearance of glucose and sucrose from the saliva of human subjects. Archs oral Biol. 30, 269-274. Stephan R. M. (1944) Intra-oral hydrogen-ion concentrations associated with dental caries activity. J. dent. Res. 23, 257-266.
Stralfors A. (1948) Studies of the microbiology ofcaries. III. The buffer capacity of the dental plaques. J. dent. Res. 27, 587-592.
Wimpenny J. W. T. (1988) Introduction. In CRC Handbook of Laboratory Model Systems for Microbial Ecosystems
(Ed. Wimpenny J. W. T.), Vol. 1, pp. 1-17. CRC Press, Boca Raton, FL.
Archs oral Biol. Vol. 37, No. I I, PP. 913-922, 199: Printed in Great Britain. All rights reserved
pH RESPONSES TO SUCROSE AND THE FORMATION pH GRADIENTS IN THICK ‘ARTIFICIAL MOUTH’ MICROCOSM PLAQUES C.
OF
H. SISSOSS, T. W. CUTRESS, G. FAULDS* and L. WOXG
HRC Dental Research Unit, Wellington School of Medicine, Otago University, P.O. Box 27-007, Wellington, New Zealand (Receiced 11 February 1992; accepted 27 Ma? 1992) Summary-Artificial microcosm plaques were grown in a five-plaque culture system for up to 6 weeks, reaching a maximum depth of several mm. Procedures for long-term pH measurement with glass electrodes were established; they showed that the application of 5 or 10% sucrose for 6 min with a slow continuous flow of a basal medium containing mucin (BMM) generated the pH changes characteristic of in rho Stephan curves. These pH responses were reproducible between plaques. Plaque mass and thickness were critical variables. Successive, sucrose-induced pH curves in plaques up to 4 mm thickness showed minor reductions only in the amplitude and rates of pH change. In plaques over 4 mm thick there was a pronounced reduction in pH response to successive sucrose applications, indicating increased diffusion limitations-a result of plaque growth to seal in the freshly-inserted pH electrode. In plaques of 6 mm maximum thickness, 10% sucrose induced a decrease to below pH 5.5 lasting 24 h, compared to the pH response in 2 mm thick plaque, which returned to the resting pH in 2 h. Differences in pH of up to 0.9 units were identified in thick plaques between inner and outer layers. The BMM flow rate was a critical determinant of the amplitude of the pH response to sucrose and subsequent return to resting pH. These results confirm, for microcosm plaque, the importance of clearance dynamics and diffusion-limited gradients in regulating plaque pH.
Key words: pH, microcosm plaque, sucrose. Stephan curve, diffusion, plaque pH gradients,
et
al., 1989: Macpherson and Dawes, 1991b; Macpherson, Chen and Dawes, 1991). It was The pH is a dominant factor in the pathology and concluded that critical variables of the film affecting ecology of dental plaque. Acidic pH fluctuations pH responses include flow rate and turnover, which in plaque induced by dietary carbohydrate have a control the supply and removal of substrates, metabtypical profile, the Stephan curve (Stephan, 1944; olites and buffers (Lagerlof, Dawes and Dawes, 1984; Kleinberg and Jenkins, 1964). Factors affecting the Dawes, 1989). and plaque thickness (Dawes and magnitude and shape of the curve include the rate Dibdin, 1986; Dawes, 1989, Dibdin 1990a, b). One approach to modelling complex microbial ecoand duration of: acidogenesis (which in turn is related to the supply, concentration and removal of carbosystems such as dental plaque is to study laboratory hydrate); alkali generation; CO, loss, removal of submicrocosms--culture systems where the mixed natural strates, metabolites and buffers into the surrounding flora is evolved in vitro under controlled environmental fluid environment, and from the mouth (Kleinberg and nutrient conditions (W5mpenny, 1988). Plaque et al., 1982; Schachtele and Jensen, 1982; Dawes, microcosms cultured in vitro from the mixed oral flora 1989; Dibdin, 1990a, b). Interactions with plaque, have advantages in investigating plaque behaviour enamel, saliva and gingival crevicular fluid buffers (Sissons, Cutress and Pearce, 1985; Sissons, Hancock are important (Dibdin, 1990b). and Cutress, 1988; Sissons et al., 1991). In order Substantial progress in quantifying the importance adequately to control and replicate experiments in of the processes involved has followed the discovery plaque metabolism and ecology it is necessary to that exchanges between supragingival plaque and the monitor, and ultimately, to manipulate plaque pH. oral environment are mediated by a thin film of saliva We have developed a five-plaque ‘artificial mouth’ (Collins and Dawes, 1987). The flow rate of this saliva microcosm for such studies in which plaques can be film and its exchange with bulk saliva vary at different grown in controlled environmental conditions, includsites (Dawes et al., 1989). Stephan pH curves resulting ing those of fluid and nutrient composition and supply. from diffusion of carbohydrate into plaque from this This provides a model system for examining the saliva film, its metabolism and end-product removal e,ffect on pH of many variables that cannot be readily have been studied in simple, in vitro models (Dawes studied in viva (Sissons et al., 1991). A continuous supply of a mucin-containing, peptone-vitamin nutrient solution, and the periodic addition of su*Present address: Department of iviolecular Pharmacology, crose simulates the in vivo dynamics of carbohydrate supply and removal, such as with the ingestion of Middlesex and Hospitals Medical School, Windeyer Building, Cleveland Street, London WIP 6BD, U.K. sucrose drinks. Because microcosm plaques model Iir(TRODL’CTION
913
C. H. Stsso~s et al.
91-t
the composition of natural plaque and hence its behaviour more closely than do simple, defined systems based on chemical exchanges or single organisms. they further test the involvement and significance of various processes in pH control. Our overall objective is to understand and control pH in plaque microcosms. Specific aims now were first to establish procedures for long-term, continuous, reproducible pH measurement in ‘artificial mouth’ microcosm plaques, and second to examine pH changes generated by transient exposure to sucrose and the effects on them of plaque thickness and of variation in the flow rate of a continuously supplied, peptone-mucin nutrient.
hI.-\TERIALS
AND METHODS
Plaque culture apparatus and growth procedures
Microcosm plaques were grown in a five-plaque artificial mouth as described by Sissons et al. (1991). It consists of: a horizontal. 110 mm dia cylinder with a detachable end-plate: five plaque-growth stations 80 mm apart; gas and thermometer ports on the top; a port in the bottom at one end leading to a 20-I waste reservoir and gas scrubbers. A later version of the culture chamber with endplates at both ends and larger side-ports was also used. Each plaque-growth statton included in a planar section of the growth chamber: opposed horizontal ports, one a B34 cone through which the intact plaque holder could be inserted; a vertical B24 port for the fluid delivery head; and angled ports on opposite sides for inoculation (B24, 60’ above horizontal) and sampling (B40, 45’ above horizontal). Reference and pH electrodes, respectively, were also installed through these angled ports. A revised fluiddelivery head was also used, which included, in addition to the fluid lines, an injection port and a drop-positioning rod to give more precise application of fluid to the plaque (Plate Fig. 1). Plaque growth was initiated by inoculation with whole mixed saliva on days 0, 3 and 5 of growth. A basal medium containing 0.25% mucin (BMM), based on the medium of Glenister et al. (1988) was supplied continuously at 3.3-3.6 ml/h to each plaque under standard conditions. It contained: 0.5% trypticase peptone, 1.0% proteose peptone. 0.5% yeast extract (all from Difco Laboratories. Detroit, MI. U.S.A.), 0.25% KCl, 5 mg/l haemin. 1 mg/l menadione, 0.25% partially purified pig gastric mucin (type III. Sigma Chemical Co., St Louis, MO, U.S.A.). In some experiments (see figure legends), the BMM was modified by supplementation with arginine (1 mmol~l) and urea (0.5, 1 or 20 mmol/l).
Every 6 or 8 h, unless otherwise stipulated, 1.5 ml of 5% w/v sucrose were supplied to each plaque over 6 min in addition to the BMM. Plaques were grown for up to 6 weeks. Experiments were made to establish the reproducible measurement of sucrose-induced pH curves, and to examine the effects of plaque mass, thickness and BMM flow rate on them. Intraplaque pH gradients during responses to sucrose were examined. Plaque support configuration
Plaques were grown in different physical configurations. The plaque supports were usually plastic, 25 mm dia. Lux Thermanox (TM) coverslips, which, for initial experiments. were attached to glass platforms as described by Sissons et al. (1991). A revised holder is now standard, where the coverslips are supported on glass rings with retaining and drainage posts [Plate Fig. 2(A)]. These arrangements allowed unconstrained plaque growth over the coverslip until the excess fell periodically to the base of the culture chamber. In addition, two plaque-support systems giving more constrained growth were developed. One was primarily for holding enamel blocks and consisted of a 16 mm square, 5 mm deep glass basket supporting either a trimmed Lux coverslip or 4 mm deep block of polyester resin (Epiglass 552P polyester encapsulated resin, Epiglass NZ Ltd, Auckland. New Zealand), usually containing four, embedded. polished, 3 x 4 mm, bovine enamel blocks [Plate Fig. 2(B)]. To examine the effect of plaque thickness. a 2.5 mm thick polyester spacer (no enamel) was placed under the trimmed coverslip to give a plaque 2.5 mm deep, as compared with a 5 mm thick plaque on a coverslip only. An approximately constant thickness was maintained by passing a Teflon blade ‘sweeper’ [Plate Fig. 2(C-iii)] across the holder. The second was a prototype constant-thickness holder [Plate Fig. 2(C-i)] consisting of a Teflon block, 36.5 mm long, 31.5 mm wide. 12.5 mm thick, threaded on to a plastic-coated steel rod. The block had a 25 mm dia hole into which tightly fitted a 10 mm thick Teflon plug with a 5 mm hole through its centre on which rested a Lux coverslip. The depth above the coverslip was set with a steel spacer [Plate Fig. 2(C-ii)] as described by Peters and Wimpenny (1988). To give drainage, four equidistant, vertical, 5 mm dia holes were drilled with their centres offset 1 mm from the rim of the central recess containing the plaque. The plaque surface was hand swept daily in the culture chamber with a 34 mm wide, 1 mm thick Teflon blade in a holder fitted with a guide to stop lateral movement [Plate Fig. 2(C-iii)], and plaque present in the drainage holes was removed.
Plate 1 Fig. 1. Fluid delivery head for plaque-growth station. This fitting is based on a B24/29 cone and is modified from the head used before (Sissons et al., 1991) as follows. The top was expanded into an approx. 30 mm dia chamber with angled tubes to take lines for sucrose and experimental treatments. an angled port closed with a Schott LC 14 cap and silicone septum for direct injection of fluids, and the main (BMM) feedline shortened to stop inside the chamber to form an infection lock. Below the B31 29 cone, the glass was shaped into a cone leading to a 4 mm dia hole (A) which was directly over a 5 mm dia bulge in the glass, 2 mm dia. drop-positioning rod. Two further holes (B and C) were positioned to allow fluid overflow and give gas-phase access if hole (A) was blocked. The end of the positioning rod was at a height of 15 mm above the Lux coverslip and positioned drops of fluid centrally over the plaque.
MicrocosmplaquepH
Plate I
C. H. Sisso?;s ei ul
Plate 2
917
Microcosm plaque pH Table I. Effect of 5 and 10% sucrose on pH change variables in plaque Variable (SD*) Initial pH pH minimum APH Minutes to pH minimum Area of curve (pH x h) Area of curve below pH 5.5 (pH x h)+
5% Sucrose 6.33 (0.13) 5.75 (0.07) -0.57 (0.07) 35.9 (8.8) 1.37 (0.20) 0
10% Sucrose
Ratio 5 IO(%)
6.32 (0.05) 5.26 (0.09) - 1.OS(0.05) 41.8 (4.5) 2.17 (0.30) 0.13 (0.06)
0.54 0.86 0.62 0
*Six replicate curves induced by 5% w Y sucrose were followed by six induced by 10% w L sucrose in :he same nlaoue, with a BX4MRow rate of 3.6mlih. +Taken as a measure of the ‘critical pH’.
pH monitoring
Reproducibility
A micro-oesophageal glass pH electrode (1.4 mm dia. Model SA3. World Precision Instruments, New Haven, CT, U.S.A.) mounted in stainless-steel tubing, and a micro-reference electrode (Model 401, Diamond General Co.. Ann Arbor. MI, U.S.A.) were installed in the culture chamber and plaque, at angles of approx. 45 and 60’ from the horizontal. They were connected through a pH meter (Model 1300. Diamond General Co.) and custom-built signal conditioner to a chart recorder (Model 8100. Philips, Eindhoven. The Netherlands). The reference electrode was positioned near the edge of the plaque and the pH electrode well into the plaque. or in various positions (see figure legends). The electrodes were calibrated periodically ivith pH 7 and 4 buffers.
Sucrose applied rapidly over 6 min (I5 ml h per plaque) to plaques grown with BMM-urea ( 1 mmol,l) arginine (I mmol/l) induced a decrease followed by a slow return to the resting pH, characteristic of Stephan curves (Text Fig. 3). In the absence of added substrates, the pH reached a stable, steady-state, resting pH that typically varied by only 0.02 pH unit over, for example, 9 h. Larger pH fluctuations sometimes occurred, probably associated with changes in plaque mass as excess plaque grew and fell from the growth station. After the insertion of the pH electrode. successive, sucrose-induced pH changes were reproducible, obtained in all but very thick plaques [Text FIN. 3(A)]. Small changes in shape of successive curves did occur in thin plaques; these were pronounced in thicker plaques [Text Fig. 3(B)]. In the 2.5 mm plaque, growth in volume was minimal, but did fill in any gap
AnalJ,tical
techniques
Plaque depth was measured lvith a periodontal probe calibrated in mm. Changes in pH were analysed using the minimum and resting pH and the maximum pH change (ApH in pH units) (Sissons and Cutress, 1987). The area of the sucrose-induced pH curve, or the area below pH 5.5 was calculated in units of pH x time (h) when appropriate.’
RESULTS Electrode
performance
After initial conditioning of a pH electrode in plaque, the slope between pH 7 and 4 remained 3 f 0.1 pH units for over 2 weeks, indicating stability of the pH electrode. The E’ (pH 7) value drifted more, a result of changes in the reference electrode. In a well-stabilized system examined after 18 days of continuous measurement, the E’ had drifted 0.08 pH unit, and the response between pH 7 and 4 had changed 0.05 pH unit. Different reference electrodes varied in stability. response and longevity.
of sucrose-induced
pH changes
caused by inserting the pH electrode. This reduced the size and rate of pH decrease and the rate of rise
by reducing access of the applied sucrose. Reproducible pH changes were obtained by inserting electrodes into plaque at daily intervals o\er 5 days [Text Figs 3(C) and (D)]. indicating that disturbing normal-thickness plaques by such insertion did not invalidate pH measurement. This allows for external electrode recalibration. In five replicate plaques grown for 2 weeks in BMM-urea (1 mmol,I)-arginine (1 mmol 1) with 5% sucrose f&hourly, the resting pH was 6.39 (SD 0.08) and the minimum pH induced by 5% sucrose was 5.83 (SD 0.10) (data not shown). Comparison
of pH responses
to 5 and 10”0 sucrose
The pH response of plaque to 10% sucrose was almost twice that of 5% sucrose (Table I). Effeect of plaque mass and thickness on sucrose -induced pH curves
The shape of sucrose-induced pH curves varied with plaque age and hence mass, and with different
Plate 2 Fig. 2. Plaque-support systems. (A) Revised, unrestrained growth, plaque-support system. The Lux coverslip rested on a 2 mm thick. 25 mm dia glass ring mounted on a glass rod with equidistant 2 mm dia glass rods raised approx. 2 mm above the ring surface to locate the coverslip and break the fluid meniscus of the plaque, and corresponding posts, reaching about 4mm below the ring, to assist in the formation of reproducible drops and hence in plaque drainage. (B) Glass-basket, plaque-support system containing four bovine enamel blocks. (C) Prototoype, constant-thickness plaque holder (i) containing a Lux (TM) coverslip, with final plaque thickness being set by the 2.0mm spacer (ii). The Teflon-blade sweeper (iii) is used to remove excess plaque.
918
C. H. SISSONSet al.
5 I
7
I
I
I
I
I
I
I
I
6) s
I
I
I
I
I
3
4
5
6
(B)
:
6 I Q 5
:q 1
1
4
0
I 12
I
I
I
I
I
I
3
4
5
6
7
12
0
Hours
1. ; !
7 1 Qa 2 m $
2.0
i? 5 0.4
1.5
1.0
& > 2
0.2
2
0.5
z
0.1
I a
0.0
: 2
0.0
12345
Day
7
Hours
Ia
0.3
5
&J, 1
12345
Day
Fig. 3. Reproducibility of sucrose-induced pH responses. (A) Four successive pH curves induced by 5% w/v sucrose (1.5 ml over 6 min, indicated by the bar labelled S) after initial pH electrode insertion into normal-thickness plaque (4mm maximum) grown in a basket holder with BMMurea (1 mmol/l)-arginine (I mmol 1) at 3.3 ml,‘h per plaque for 4 weeks with 6-hourly 5% w, v sucrose. The substratum was polyester resin containing four bovine enamel blocks. (B) Successive, sucrose-induced pH curves after pH electrode insertion into thick plaques. and effect of plaque mass. Plaques grown for 25 days with BMM-urea (1 mmol/l)arginine (I mmol/l) at 3.3 ml/h per plaque and 6-hourly 10% w/v sucrose were mixed and deposited to 2.5 mm thickness on a trimmed Lux coverslip in a basket holder. The plaque mixture was grown for 4 days with I-hourly 10% w/v sucrose, the surface trimmed and the pH electrode inserted near one corner. pH changes were induced by d-hourly 10% w/v sucrose, 1.5ml over 6 min as shown by the bar. After the first sucrose-induced pH curve (2 x PLQ, . . .) approx. 50% w/v of the plaque was removed from the area distant from the pH electrode, leaving the plaque around it undisturbed and the same thickness. The following sucroseinduced pH curve (30min after plaque removal) is shown (1, -), and curves 24 h (2, ---) and 48 h (3, -) later. During this period the plaque mass was not visibly increased. (C) The pH decrease (ApH) induced by 5% w/v sucrose (1.5ml in 6 min) over 5 days, measured by daily reinsertion of electrodes into the same plaque. The mean pH decrease was 1.31 (SD 0.11) pH units. The plaque had been grown for 12 davs in BMM-urea (1 mmol’l)-arginine (1 mmol/l) at 3.3 ml/h per plaque on enamel in polyester resin (as in A). (D) Areas of sucrose-induced pH curves during the 4 h after sucrose addition for the pH decreases shown in (C). The mean area was 0.296 (SD 0.033) pH x h.
0
10
r, 20 30 Hours
40
!I
Fig. 4. Effect of plaque thickness on sucrose-induced pH changes. (A) Sucrose-induced pH changes in plaque at different thickness (and growth stage) grown in BMM-urea (0.5 mmol/l)-arginine (I mmol/l), 3.6 ml/h per plaque. Five % sucrose was applied, 1.5 ml over 6 min, as shown by the bar labelled S. Curve I (-) shows the pH changes in a 2-weeks, approx. 4 mm plaque. Most plaque was removed except for a thin layer and curve II c ,) shows the pH curve after 2 days of regrowth to approx. 1.5 mm thick. (B) Representative pH response to 5% w/v sucrose (1.5 ml over 6 min at the time shown by the bars) in a plaque grown for 27 days in BMM (3.6ml/h per plaque) whilst constrained to constant thickness (2.5 mm) by being swept daily (-). Also shown is a plaque grown for the same time but with unrestrained growth to approx. 4 mm maximum thickness (---), (C) pH changes induced by 10% w/v sucrose in a very thick (6 mm maximum) plaque. The plaque was 33 days old, grown on a 25 mm dia Lux coverslip mounted on a glass-platform holder in BMM for 2 weeks and then in BMM-urea (20 mmol/l) at 3.6 ml/h for the remainder of the period. The pH electrode was positioned centrally in the plaque 2 weeks previously. The arrow indicates time of 10% w/v sucrose addition (1.5 ml in 6 min).
types of plaque holder. Plaques grown unrestrained on coverslips were convex and generally increased in size to a maximum that depended, for example, on the impact of the drops of fluid supplied and on shelter by electrodes. Here, sucrose-induced pH curves ranged from a rapid decrease and return to the resting pH in freshly grown plaques, to a slower, smaller decrease that required over 4 h to return to resting pH in thicker and older plaques [Text Fig. 4(A)].
919
Microcosm plaque pH
I
n
1: I
;
1
I
I
0
12
I
I
I
I
I
I
3
4
5
6
7
Ia
I
I
I
I
I
I
I
I
I
_
_p)
6 qy-
Hours 5 (6)
3tV1 0
12
6.0
I 3
I 4
I
I
I
5
6
7
6
Hours
r-l % 0
0
-7&h!&-+
BMM flow rate (ml/h/plaque)
Fig. 5. The effect of BMM flow rate on sucrose-induced pH changes. pH changes induced by 10% w/v sucrose (1.5 ml over 6 min. shown by the bar labehed S) in a plaque (3-4 mm maximum thickness) previously grown on BMM for 9 days on a 25 mm coverslip (glass-platform holder) were examined over 5 days at BMM flow rates of 0, 1.8, 3.6, 7.2 and 18 ml/h per plaque. These different flow rates diluted the sucrose to 10.0,8.9,8.1,6.8 and 4.6%. Replicate curves were examined at intermediate flow rates. (A) Representative pH changes at BMM flow rates of 0, 1.8, 7.2 (-, as labelled) and 18 ml/h ( .). (B) The sucrose-induced minimum pH at different flow rates.
Removal of half the plaque mass increased the amplitude and rates of pH change [Text Fig. 3(B)]. When the growth of plaques to a constant size and thickness was examined in a constant-thickness holder where plaque above 2.5 mm was removed daily [Text Fig. 4(B)], the pH decrease in plaque of constant 2.5 mm thickness was greater than in similarly aged plaques with unrestrained growth (4-5 mm, maximum depth). In another experiment, pre-grown plaque was deposited to give 2.5 and 5mm thicknesses in the baskets designed for experiments on enamel [Plate Fig. 2(B)]; 8 h after sucrose exposure, the 5 mm plaque was still pH 4.6 whereas the 2.5 mm plaque had returned from a pH 5.0 minimum to its resting pH of 6.4. Very thick plaques (e.g. 7 mm maximum depth) formed on the 25 mm dia Lux coverslips after several weeks of growth could give long-lasting pH changes. For example, in a 6mm plaque, the pH changes induced by 10% sucrose [Text Fig. 4(C)] had a minimum lasting 15 h below pH 5, and took 2 days to return to the resting pH, despite the presence of 20mmol/l urea in the BMM (which had raised the resting pH to 7).
Fig. 6. pH gradients formed during sucrose metabolism. Plaque was grown in BMM-urea (1 mmol/l)-arginine (1 mmol/l) at 3.3 ml/h per plaque on 25 mm dia Lux coverslips. It reached approx. 7 mm depth in 16 days. The pH electrode was located in the top surface layers (T) of the plaque or at the base (B). The bar shows 5% w;‘v sucrose addition (1.5 ml over 6 min). (A) Five % sucrose-induced pH changes at the base and surface of the plaque. For the curve labelled B (-), the pH electrode was inserted deep to the plaque base 12 h before the addition of sucrose. The curve labelled T (-) shows the second sucrose-induced pH curve in which the pH electrode was located in the plaque surface layers. (B) The pH response measured with the pH electrode repositioned between the plaque surface and the plaque base before and during the pH fluctuation. These sucrose-induced pH responses were flanked by the pH response curves shown in (A), curve B which is shown as a dashed line. The pH electrode was positioned initially at the base of the plaque at point l(B). The pH electrode was raised to the top surface layers of the plaque, point 2(T). The surface pH was recorded and then sucrose added. After the rapid surface pH fall during sucrose application [to point 3(T)], the pH electrode was inserted to the base of the plaque. The higher base pH was measured at point -I(B’). The pH electrode was then repositioned in the plaque surface, point 5(T’), where it measured the rise portion of the Stephan curve in the plaque surface layer.
Effect of BMMflow rate on sucrose-inducedpH
curves
Limiting the flow rate of BMM in normal (2-4 mm maximum thickness) plaques had a substantial effect on the Stephan curve (Text Fig. 5). If the supply of BMM was stopped at the time of sucrose addition, the pH fell and remained low [Text Fig. 5(A)]. Increasing the BMM flow rate reduced the amplitude of the pH change until further increases in flow rate made little difference. The minimum pH reached in the Stephan curve proved to be the most replicable measure of the pH changes in these experiments. It increased with increasing BMM flow rate to a maximum value [Text Fig. 5(B)]. The initial pH was largely a function of the shape and duration of the previous sucrose-induced pH curve and varied significantly during these experiments on pump-rate variation. Differences in initial pH invalidate direct comparisons with measures such as the magnitude of the pH decrease and the area of the pH curve. Plaque pH gradients
The growth of plaques several mm thick provided an opportunity to examine intraplaque pH gradients.
920
C. H. S~sso~set al.
AS fully developed plaques on 25 mm dia coverslips neared the resting pH. there was a small but significant 0.1-0.2 pH unit gradient from the edge to the centre of the plaque at the base. However, in a thick (7 mm maximum) plaque in response to metabolism of 5% sucrose, substantial pH gradients were formed between the top surface of the plaque centre and its base. After sucrose application, the pH response on the plaque surface was a typical Stephan curve compared with virtually no pH change in deep plaque against the coverslip [Text Fig. 6(A)]. This pH gradient was also demonstrable by moving the pH electrode between the plaque surface and base during a Stephan curve [Text Fig. 6(B)]. The electrode in the plaque surface during sucrose application showed a sucrose-induced decrease at the surface of 0.95 pH unit. Inserting the electrode to the base of the plaque indicated that there was a sucrose-induced decrease of 0.06 pH unit only. A comparison with Text Fig. 6(A) shows that moving the electrode caused some mixing of surface and base plaque pH conditions. DISCUSSION
Long-term
plaque pH measurement
No ideal procedure for the long-term measurement of plaque pH in situ has yet been devised. Calibration,
drift. and the response of the measurement system to the plaque environment are issues that need to be addressed in in vitro (Russell and Coulter, 1975; Hudson, Donoghue and Perrons, 1986) and intraoral studies (Kleinberg et al., 1982; Schachtele and Jensen. 1982; Igarashi, Lee and Schachtele, 1989, 1990) of plaque pH. In cico experiments are limited to a few days and give little possibility for controlling plaque formation or environment. In clitro experiments can be extended to many weeks, but calibration problems are increased. The drift in the response of our electrode system, though small, necessitated occasional recalibration. Periodic removal of the electrodes and calibration checking has the advantage of confirming the reliability of the system overall. Limitations include disruption of plaque structure and difficulty in accurately positioning the pH electrode. In thick plaques disturbed by inserting electrodes, the size and rate of pH decrease with successive sucroseinduced pH changes indicates increased diffusion barriers as a result of plaque growth. Nevertheless, the pH cumes measured were relatively reproducible, especially considering that plaque morphology and mass were usually undefined. .\ficrocosm plaques natural plaque
as a model for pH regulation
in
Studies of carbohydrate-induced, plaque pH changes involving computer and simple, in vitro, monobacterial models have indicated that limitations on diffusion set up interacting intraplaque gradients of pH, acid- and alkali-generating substrates, acids, ammonia and mobile buffers (Dawes and Dibdin, 1986: Dawes, 1989, Dibdin, 1990a, b; Macpherson and Dawes, 1991a, b; Macpherson et al., 1991). Plaque pH is determined by the supply, metabolism and clearance dynamics of these compounds, and their
interaction with bacterial cell buffers and enamel. Plaque thickness, and the rate of supply of oral fluids, generally saliva, are major variables affecting the outcome. Microcosm plaques are distinct from defined bacterial systems. They are similar in appearance, structure. composition and behaviour to natural plaques (Sissons et al., 1991); our results confirm this for pH behaviour in response to sucrose. Variables critical to pH control can be studied in microcosm plaques with unrestrained growth, and potentially studied in plaques of more defined size and shape. A sucrose rinse lasting a few minutes, followed by sucrose and metabolite clearance, simulates the processes that arise in riro during the intake of sucrose drinks (Lagerlof et al., 1984; Sreebny, Chatterjee and Kleinberg, 1985: Lindfors and Lagerlof, 1988). This rinse is commonly used in modelling cariogenic challenges (Dawes, 1989; Igarashi et al., 1989; 1990). The pH changes typical of Stephan curves in the microcosm plaques were usually of longer duration than the curves described for smooth-surface enamel plaque (e.g. Kleinberg et al., 1982; Schachtele and Jensen, 1982; Igarashi et al., 1989. 1990). presumably because of the size of the microcosm plaques. The increased magnitude of the pH response to 10% sucrose compared to 5% is similar to that predicted in computer models (Dibdin, 1990a. b). reported in model systems (Lagerlof et al., 1984), and for in riro plaques (Lindfors and Lagerlof, 1988). Replicate plaques Fave a highly reproducible resting pH that was raised If urea was present in the BMM (Sissons et al., 1991). as shown in Text Fig. 4(C). Replication of sucrose-induced pH curves in replicate plaques was good; any variation can be explained by a sensitive dependence on the stage of plaque growth and its size. Minor variations in drainage seemed to affect the pattern of plaque growth and would potentially affect intraplaque gradients, clearance of substrates, metabolites, and hence pH. One reason that the minimum pH was the most repeatable measure of the sucrose-induced pH curve is that the pH scale is a logarithmic transform of hydronium ion concentration, which would tend to minimize experimental variation at low pH. This may be countered to a degree by the high buffering capacity of plaque (Stralfors, 1948; Sissons and Cutress, 1987, Shellis and Dibdin, 1988; Dibdin, 1990b; Macpherson and Dawes, 1991b), which would tend to make the pH response linear. Plaque sire, thickness and Stephan curces
The potentially profound effect of plaque thickness on the Stephan pH response has been indicated by computer modelling (Dawes and Dibdin, 1986; Dawes, 1989; Dibdin, 1990a, b). The thickness of natural dental plaque varies with location. On smooth enamel surfaces it ranges up to 200pm (Main et al., 1984). However, in fissures and approximal sites, plaque is potentially l-2 mm thick as measured from the surface exposed to saliva. These are sites of high caries susceptibility. Intra-oral models of fissure plaque have ranged up to 2 mm depth (Igarashi et al., 1989). There seems to be little published information on the size and thickness of approximal plaques but where oral hygiene is poor, and tooth or restoration configuration appropriate, they potentially grow to
921
Microcosm plaque pH
virtually the full width of molars. The inner regions of such plaques are inaccessible to saliva exchanges, and Stephan curves in them can remain at a low pH for long periods (Schachtele and Jensen, 1982; Igarashi et al., 1989, 1990). Microcosm plaques grow up to several mm thick. Those of 6mm thickness, with sucrose-induced pH curves below pH 5 for over 12 h, probably reflect a more extreme situation than occurs in vivo. They provide an opportunity to study pH behaviour at the limits of diffusion restrictions and mode! this aspect of approximal and fissure plaques. The size and shape of plaques in the mouth are constrained by movements of the tongue, lip and cheeks, by mastication and by oral hygiene procedures. It would be desirable to simulate these processes in vitro to produce plaques of defined morphology. A daily sweep with a Teflon blade over a 25 mm dia, constant-thickness plaque holder defined plaque thickness and resulted in sucrose-induced pH curves of greater amplitude than in free-growing plaques. How often the actively growing plaque surface is removed will affect the selective pressures on plaque growth. Allowing fluid drainage from the plaque proved important in achieving a relatively solid biofilm. It avoided the submersion of plaque in a substantial liquid phase (Peters and Wimpenny, 1988) and increased the clearance of soluble substrates and metabolites including toxins. More in vitro and in riro research is needed before it will be possible to simulate realistically in citro the processes that physically limit plaque in ciro. BMM jlow rate and Stephan curves Applying sucrose at 15 ml/h per plaque ensured that there was only modest dilution by the BMM at normal flow rates, resulting in a limited effect on the rate of pH decrease during sucrose application. Subsequently the BMM flow rate set the clearance characteristics of the system and proved to be of critical importance in modifying the shape of the Stephan curves. With no flow at al!, the pH decrease was large and the pH did not rise from the minimum. At higher flow rates the pH decrease was very small. Under our standard growth conditions, the BMM flow rate of 3.3-3.6 ml/h per plaque limited the clearance dynamics. The great variation in total pH decrease and curve shape with flow rate (Text Fig. 4) confirms, for microcosm plaques, the dependence of plaque pH behaviour on clearance dynamics into the oral fluid, as demonstrated in simple chemical (Dawes et al., 1989), computer (Dawes and Dibdin, 1986; Dibdin 1990a, b) and monobacteria! mode! systems (Macpherson and Dawes, 199!a, b; Macpherson et al., 1991). Intraplaque
pH gradients
Dawes and Dibdin (1986) predicted from computer modelling based on diffusion theory that restricted saliva supply dynamics to plaques results in diffusionlimited, intraplaque gradients that regulate the pH. This now has been eIegant!y demonstrated in flowcontrolled, thin liquid-film models based on biofilms of bacteria suspended in agarose (Macpherson and Dawes, 199la, b; Macpherson et al., 1991). There
are also steep oxygen gradients in plaque. such that anaerobiosis is likely within a few hundred pm of a plaque surface exposed to atmospheric oxygen concentrations (Globerman and Kleinberg, 1979). To the best of our knowledge the 0.9 pH unit gradient resulting from a substantial 5% sucroseinduced pH response on the surface of a thick microcosm plaque and the very small pH fluctuation in inner plaque constitute the first reports of large pH differences in a system which approximates to that of natural plaques in composition. This further confirms the existence in dental plaque of the pH gradients that were demonstrated in the studies on Stephan curves in Streptococcus oralis biofilms in agarose (Sfacpherson and Dawes, 199lb; Macpherson et al., 1991). These showed that when the flow of fluid film over plaque is restricted. pH gradients down and along the ‘plaque’ reach over 2 pH units with 10% sucrose. Although in normal-thickness microcosm plaques, 10% sucrose increased the amplitude of the sucrose-induced pH curve (Table 1), in very thick plaques (despite diffusion barriers), the sucrose clearly penetrated sufficiently to lower the pH. However, diffusion limitations reduced the clearance of acid products and thus prolonged the pH minimum and retarded the return to resting pH. A major reason for pH gradients and slow pH responses is the high capacity of plaque cell buffers (Stralfors. 1948; Sissons and Cutress, 1987; Shellis and Dibdin, 1988: Dibdin and Shellis, 1989; Dibdin, 1990a, b; Macpherson and Dawes, 199lb. Macpherson et al., 1991). Whilst diffusion of free molecules may be little restricted in plaque (Dibdin, 1990a, b), they interact with, and have to shift the ionization state of, an approximately 500-fold excess of fixed ionic groups (Sissons and Cutress. 1987). In addition to extracellular stationary and mobile buffers, cell transport processes and intracellular buffering systems may be important in regulating plaque pH. Acknowledgements--We thank E. M. Hancock and P. Smith for technical assistance. J. Teppett, Wellington Public Hospital. designed and constructed the signal conditioner. S. Newcombe and K. McCombe. Victoria Cniversity of Wellington, advised on and did the glass-blowing for the modified culture chamber and fittings. We gratefully acknowledge financial support for this project from L’Oreal Laboratoires de Recherche Appliquee et Dtveloppment, Clichy, France.
REFERESCES
Collins L. M. C. and Dawes C. (1987) The surface area of the adult human mouth and thickness of the salivary film covering the teeth and the oral mucosa. J. dent. Res. 66, 1300-1302. Dawes C. (1989) An analysis of factors influencing diffusion from dental plaque into a moving film of saliva and the implications for-caries. J. dent. kes. 68, 1183-1488. Dawes C. and Dibdin G. H. (1986) A theoretical analvsis of the effects of plaque thickness and initial salivary sucrose concentration on diffusion of sucrose into dental plaque and its conversion to acid during salivary clearance. J. dent. Res. 65. 89-94. Dawes C., Watanabe S., Biglow-Lecombe P. and Dibdin G. H. (1989) Estimation of the velocity of the salivary film at some different locations in the mouth. J. denr. Res. 68, 1479-1482.
922
C. H. SISSO?;S er al.
Dibdin G. H. (1990a) EXect on a cariogenic challenge of saliva/plaque exchange via a thin salivary film studied by mathematical modelhng. Curies Res. 24: 231-238. Dibdin G. H. (1990b) Plaaue fluid and diffusion: studv of \ the cariogenic challenge by computer modelling. J. ient. Res. 69, 1324-1331. Dibdin G. H. and Shellis R. P. (1989) The interpretation of CO, equilibration data to obtain plaque fluid buffer capacities- and comparison with results obtained by titration. J. dent. Res. 68. 1323-1327. Glenister D. A., Salamon Kl E., Smith K. and Beighton D. (1988) Enhanced growth of complex communities of dental plaque bacteria in mucin-limited continuous culture. I
I
Microbial. Ecol. Hlth Dis. 1, 31-38.
Globerman D. Y. and Kleinberg I. (1979) Intra-oral p0, and its relation to bacterial accumulation on the oral tissues. In Sulica and Dental Curies (Eds I. Kleinberg, S. A. Ellison and I. D. Mandel), pp. 275-291. Microbioloy Abstracts, Information Retrieval Inc.. New York. Hudson D. E., Donoghue H. D. and Perrons C. J. (1986) A laboratory microcosm (artificial mouth) for the culture and continuous pH measurement of oral bacteria on surfaces. J. uppl. Bucrerio[. 60, 301-3 10. Igarashi K., Lee I. K. and Schachtele C. F. (1989) Comparison of in uiro human dental plaque pH changes within artificial fissures and at interproximal sites. Curies Res. 23, 417-422. Igarashi K., Lee I. K. and Schachtele C. F. (1990) Effect of dental plaque age and bacterial composition on the pH of artificial fissures in human volunteers. Curies Res. 24, 52-58.
Kleinberg I. and Jenkins G. N. (1961) The pH of dental plaques in the different areas of the mouth before and after meals and their relationship to the pH and rate of flow of resting saliva. Archs oral Biol. 9, 493-516. Kleinberg I., Jenkins G. h’.. Chatterjee R. and Wijeyweera L. (1982) The antimony electrode and its role in the assessment and interpretation of dental plaque pH. J. dew Res. 61, 1139-I 147. Lagerlof F., Dawes R. and Dawes C. (1984) Salivary clearance of sugar and its effects on pH changes by Sueptococcus mirior in an artificial mouth. J. dent. Res. 63, 1266-1270. Lindfors B. and Lagerlof F. (1988) Effect of sucrose concentration in saliva after a sucrose rinse on the hydronium ion concentration in dental plaque. Curies Res. 22, 7-10. Macpherson L. M. D. and Dawes C. (1991a) Urea concentration in minor mucous gland secretions and the effect of salivary film velocity on urea metabolism by Slreptococcus cesfibularis in an artificial plaque. J. periodont. Res. 26, 395-401. Macpherson L. M. D. and Dawes C. (1991b) Effects of salivary film velocity on pH changes in an artificial plaque
containing Srreptococcus oralis, after exposure to sucrose. J. dent. Res. 70, 1230-1234. Macpherson L. M. D., Chen W. Y. and Dawes C. (1991) Effects of salivary bicarbonate content and film velocity on pH changes in an artificial plaque containing Sfreptococcus oralis, after exposure to sucrose. J. dent. res. 70, 1235-1238. Main C., Geddes D. A. M., McNee S. G., Collins W. 1. N., Smith D. C. and Weetman D. A. (1984) Instrumentation for measurement of dental plaque thickness in situ. J. biomed. Eng. 6, 151-154. Peters A. and Wimpenny J. W. T. (1988) A constant-depth laboratory model film fermenter. In CRC Handbook of Laboratory Model Systemsfor Microbial Ecosystems (Ed. Wimpenny J. W. T.), Vol. 1, pp. 175-195. CRC Press, Boca Raton, FL. Russell C. and Coulter W. A. (1975) Continuous monitoring of pH and Eh in bacterial plaque grown on a tooth in an artificial mouth. Appl. Microbial. 29, 141-144. Schachtele C. F. and Jensen M. E. (1982) Comparison of methods for monitoring changes in the pH of dental plaque. J. dent. Res. 61, 1117-l 125. Shellis R. P. and Dibdin G. H. (1988) Analysis of the buffering systems in dental plaque. J. dent. Res. 67, 438-446.
Sissons C. H. and Cutress T. W. (1987) In-&o ureadependent pH-changes by human salivary bacteria and dispersed, artificial-mouth, bacterial plaques. Archs oral f&l. 32, 181-189. Sissons C. H.. Cutress T. W. and Pearce E. I. F. (1985) Kinetics and product stoichiometry of ureolysis by human salivary bacteria and artificial-mouth plaques. Archs oral Biol. 30, 781-790.
Sissons C. H., Hancock E. M. and Cutress T. W. (1988) The source of variation in ureolysis in artificial plaques cultured from human salivary bacteria. Archs oral Biol. 33, 72 l-726.
Sissons C. H., Cutress T. W., Hoffman M. P. and Wakefield J. St J. (1991) A multi-station dental plaque microcosm (artificial mouth) for the study of plaque growth, metabolism, pH and mineralization. J. denl. Res. 70, 14091416. Sreebny L. M., Chatterjee R. and Kleinberg I. (1985) Clearance of glucose and sucrose from the saliva of human subjects. Archs oral Biol. 30, 269-274. Stephan R. M. (1944) Intra-oral hydrogen-ion concentrations associated with dental caries activity. J. dent. Res. 23, 257-266.
Stralfors A. (1948) Studies of the microbiology ofcaries. III. The buffer capacity of the dental plaques. J. dent. Res. 27, 587-592.
Wimpenny J. W. T. (1988) Introduction. In CRC Handbook of Laboratory Model Systems for Microbial Ecosystems
(Ed. Wimpenny J. W. T.), Vol. 1, pp. 1-17. CRC Press, Boca Raton, FL.
