SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

T. D. MORGAN, A. E. BEEZER, J. C. MITCHELL AND A. W. BUNCH. 2001.

Aims: This study aimed to evaluate the efficacy of ‘natural’ putative antimicrobial agents against Streptococcus mutans and to compare these with synthetic agents using the flow microcalorimeter. Streptococcus mutans is one of the oral pathogens responsible for dental caries.

Methods and Results: Traditional microbiological techniques are invasive and destructive unlike flow microcalorimetry. This rapid technique was used to continuously monitor the power output (bioactivity) of Strep. mutans with reproducibility, precision and accuracy. The antibacterial agents found in oral hygiene products and all the natural agents tested showed anti-Strep. mutans ability.

Conclusions: In this study microcalorimetry identified agents that had a biological effect and quantified the rate of kill achieved enabling four broad categories of antimicrobial agent to be defined.

Significance and Impact of the Study: Microcalorimetric data are a better indication of antimicrobial efficacy than merely determining concentrations at which an antimicrobial agent is bacteriostatic or bactericidal.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

The two most common types of dental disease, dental caries and periodontal disease, are plaque-related infections. Dental caries involves demineralization, cavitation and breakdown of calcified dental tissue and is caused by microorganisms that ferment dietary carbohydrates, notably sucrose, to produce acids; these acids initiate dissolution of the tooth enamel (Hardie 1992). Strains of Streptococcus mutans have long been implicated in the formation of dental plaque and cariogenicity (Clarke 1924; Marsh and Martin 1992). Whilst antimicrobial mouthrinses reduce plaque on visible tooth surfaces their ability to penetrate between the teeth is not sufficient to have an effect on interdental plaque where dental floss and interdental cleaners are required (Finkelstein et al. 1987). Clinical applications for antimicrobial mouthrinses fall into three broad categories: preventative, therapeutic and professional (Fischman 1994). Mouthrinses often contain surfactant, alcohol, fluoride and antimicrobial agents. The antimicrobial agents will ideally have the following properties: broad spectrum antimicrobial efficacy, retain antimicrobial efficacy at low concentrations, fast acting, non-toxic, non-irritant, pleasant/neutral odour and taste, good oral retention properties, not too disruptive to the oral microbial ecology, globally regulatory approved, chemically defined, chemically stable, physically stable and cost effective.

Natural products are now preferred by a large proportion of the population and have been reported to possess antimicrobial activity (Taniguchi and Kubo 1993; Hammer et al. 1999). Phytochemicals have recently been shown to be a good alternative to synthetic chemical substances for caries prevention (Kubo et al. 1992; Kubo et al. 1993a; Muroi and Kubo 1993a; Hamilton-Miller 1995). Extracts from Celastrus scandens, Chamaebatia foliolosa, Digitaria sanguinalis, Ginkgo biloba, Juniperous virginiana (Heisey and Gorham 1992), Anacardium occidentale L. (Muroi and Kubo 1993b) and Ilex paraguayensis St. Hil., or mate tea (Kubo et al. 1993b), have proved to be effective against Strep. mutans.

One of the current strategies is to use antimicrobial agents at a minimum yet clinically effective level as consumers demand fewer synthetic chemicals in less quantity. Agents delivered from dentifrices have a relatively short half-life in the mouth and may be present for considerable periods at sub-minimum inhibitory concentration (sub-MIC) levels, but still be of benefit because they repress the metabolism of micro-organisms (Marsh 1992). The microcalorimeter may be regarded as a bioactivity monitor that can detect changes in the metabolism of a population of micro-organisms and is ideal to quantify sub-MIC interactions with micro-organisms. This mode of action would prevent rapid changes in the oral environment and, therefore, provide stability of the microflora (Marsh 1994). The reproducibility, precision and accuracy attainable by flow microcalorimetric investigations (Chowdhry et al. 1983) of the interaction of an antimicrobial agent with micro-organisms make it possible to establish the biological response at particular antimicrobial concentrations. Microcalorimetric data during growth and inhibition of micro-organisms provide stoichiometric and energetic information about various features of metabolic activities (Roels 1983) which, following calibration with classical microbiological techniques, can be interpreted in terms of bacteriostatic or bactericidal events.

In this study we show how microcalorimetry can be used to initially screen natural products for potentially useful bioactive properties for use in mouthrinse formulations.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Organism and growth conditions

Streptococcus mutans NCTC 10832 was grown overnight in tryptone soya broth, then harvested by centrifugation (4000 g), washed and resuspended in quarter-strength Ringer’s solution containing 10% dimethylsulphoxide. Aliquots (2·01 ml) of each species were stored at − 196 °C in screw-capped cryogenic ampoules. For use, frozen cell suspensions were thawed in a 40 °C water-bath for 2 min and vortexed for 1 min. The 40 °C temperature of the water-bath was high enough to completely thaw the frozen aliquots in the ampoules.

Microcalorimetry

A flow calorimeter (LKB 10700-1; Thermometric AB, Bromma, Sweden) was linked to a null detector microvoltmeter (155; Keithley Instruments GMBH, Germany) to amplify the signal. All experiments were undertaken in a constant temperature room (21±0·1 °C), the calorimeter unit being in a thermostatted air-bath set at 30 °C. A bubble of air was allowed to enter the flow system before an experiment; this showed the boundary between the liquids, acted as a marker and also helped to clean the reaction vessel before the new liquid entered. A volume (24·6 ml minus the volume of antibacterial agent to be added) of phosphate buffer (16 mmol l−1), pH 7·2, with 12·2 mmol l−1 glucose and 20% v/v bovine heart infusion (BHI) broth was continuously stirred outside the microcalorimeter while being pumped at a constant rate through the microcalorimetric reaction vessel. When the closed loop was formed, 0·4 ml Strep. mutans were added to the outer stirred vessel and, following a 3-min lag time, the power output of the bacteria was registered.

Antibacterial agent assessment

Each putative antibacterial agent, dissolved in an aqueous solution (supplied by Johnson and Johnson, Southampton, Hampshire, UK), was tested at varying concentrations; 20 min after inoculation, one antimicrobial agent was added to bring the final volume in the outer vessel to 25 ml. Any decrease in power output from this stable upper baseline was therefore the ‘biological response’ of Strep. mutans as a consequence of the added agent. This ‘response’ was corrected for the dilution effect of adding a defined volume of liquid. The natural antibacterial agents were crude aqueous extracts and were not modified before the microcalorimetric studies. The aqueous extractant solution alone was also tested.

Log dose response

The percentage biological response (biological response divided by the difference between the upper and lower baselines) of the bacteria plotted against the logarithm of the dose of antibacterial agent applied formed a sigmoidal-shaped curve. The linear portion of the typical sigmoidal log dose response (LDR) curve may be extrapolated to where the biological response = 0 to find the maximum dose at which no biological response was elicited. This gives the artificial value, log(dose)max, that aids ranking of the antimicrobial agents in order of potency of unit dose.

inline image

The dose of the antimicrobial agent is expressed as: a concentration (mol l−1), a volume (ml) per 25 ml reaction volume (v/v) or a weight (g) per 25 ml reaction volume (w/v). The LDR curve for each extract was prepared on a consistent basis (the same concentration dimension was used).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Microcalorimetric power output against time

Figure 1 shows a stylized trace from a typical microcalorimetry experiment as performed using the protocol described earlier. Power output reached a steady state of mean 70 ± S.D. 5·4% from previously cryopreserved cells; this equates to a 97% recovery of viable cells originally present (Morgan and Bunch 2000). No decline in power output from the steady upper baseline was seen during a 4h extended run. When an antimicrobial agent was added to the flow medium, the drop in power output (biological response) was then measured. All the experiments performed resulted in a similar pattern of heat generation.

image

Figure . 1. The microcalorimetric deflection against time. The lower baseline was formed from the power output when sterile medium was pumped through the microcalorimeter in a closed loop. After inoculation with 0·4 ml Streptococcus mutans the power output from the bacteria was measured and an upper baseline was formed. The addition of a particular concentration of an antibacterial agent leads to a deflection back towards the lower baseline which is called the biological response

Download figure to PowerPoint

Log dose response curves

The response of Strep. mutans to a range of antibacterial agents at varying concentrations was tested. Plotting the biological responses against their respective logarithm of the doses of the antimicrobial agent applied produced an LDR curve where the log(dose)max was the logarithm of the dose that can be applied without eliciting a biological response. The log(dose)max values calculated for the range of antimicrobial agents studied are shown in Table 1. Morgan and Bunch (2000) reported typical responses from both sodium fluoride and Coridothymus capitatus (thyme) extract. The addition of potential active agents can cause a biological response, but may not reduce the power output to the lower baseline. Tabulation of the values of the gradient of the LDR curve (percentage biological response after 15 min per unit dose) and the log(dose)max for each antimicrobial agent studied allowed comparison of the dose effect.

Table 1.  The concentration range at which the antimicrobial agents were tested and the calculated dose and log dose response (LDR) gradient values Thumbnail image of

Comparing natural and synthetic compounds

Given the criteria of the ideal properties of antimicrobial agents to be included in mouthrinses outlined earlier, the results indicated that antimicrobial agents are required to be fast acting. The log(dose)max 10 was determined from data at 10 min after the addition of antimicrobial agent while the log(dose)max 15 was determined from data at 15 min after the addition of antimicrobial agent. An antibacterial agent that does not achieve its full effect in the first 10 min after inoculation will therefore have a lower log(dose)max 10 value than the log(dose)max 15 value. Subtracting the log(dose)max 10 from the log(dose)max 15 and comparing this result (Table 1) with zero gave an insight into the rate of action of the antimicrobial agent.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

The character and extent of the microcalorimetrically measured biological response of bacteria to an antibacterial agent was reproducible for each agent studied. The MICs, agar diffusion assays or impedance/conductance technique could be used to assess the activity of putative antimicrobial agents. Microcalorimetry, however, can indicate a wider range of metabolic events than merely percentage inhibition. The microcalorimetric determination of antibacterial efficacy requires fewer than 60 min to quantify whereas classical microbiological techniques could need a period of 48 h growth to form a visible colony-forming unit on an agar plate. It is possible to add mixtures of phytochemicals to the formulation components used in mouthwashes using microcalorimetry and it is a quicker method of detecting synergistic or antagonistic interactions than classical microbiology. There are four broad categories that can be discerned, from this study, in the comparison of the dose effect.

i Category 1. High LDR gradient, high log(dose)max. A large concentration of antimicrobial agent was required for any antimicrobial action, but a small increase in concentration had a great effect on improving antimicrobial efficacy.

ii Category 2. High LDR gradient, low log(dose)max. Desirable antimicrobial agents exhibited both of these properties. Only a small concentration was required for antimicrobial action and a small increase in concentration improved antimicrobial efficacy.

iii Category 3. Low LDR gradient, high log(dose)max. A large concentration of antimicrobial agent was needed for any observable antimicrobial action. Increased dosage did not give a corresponding significant increase in antimicrobial efficacy.

iv Category 4. Low LDR gradient, low log(dose)max. Desirable antimicrobial agents also exhibited both of these properties. A small concentration of antimicrobial agent was required for antimicrobial action and, on dilution of the agent, antimicrobial action persisted.

Dilution of mouthwash with saliva in the mouth would decrease the concentration of antimicrobial compounds; agents with a low LDR gradient and a low log(dose)max would provide continued effectiveness. From the categories described above, the even-numbered categories show better antimicrobial properties than those that are odd numbered. A good antibacterial agent such as cetyl pyridinium chloride (CPC) has a low log(dose)max while a poor antibacterial agent such as ethanol has a log(dose)max approaching zero. The gradient of the LDR curve is independent of the units qualifying the dose concentration; the log(dose)max is not as selectively neutral.

Although irrigating the gingivae with water has a beneficial effect on gingival health, water does not have the antimicrobial properties required to reduce the oral microbial flora and amount of plaque (Perdok et al. 1990) other than the loss of micro-organisms purely by water irrigation. Mouthrinses therefore often contain a number of antibacterial agents. Sodium fluoride appears as a category 2 compound. The antimicrobial efficacies, against Strep. mutans, of a number of compounds from synthetic and natural sources have been reported. Those compounds found in oral hygiene products were shown to have good anti-Strep. mutans efficacy. Chlorhexidine appears as a category 4 compound. CPC appears as a category 2 compound and had the lowest (best) log(dose)max tested. Table 1 shows CPC and chlorhexidine to have far superior antimicrobial efficacy over triclosan (a category 2 compound), highlighting the need to study antimicrobial efficacy in vivo where substantivity plays a major role.

The synthetic antimicrobial agents have low log(dose)max values and the natural antimicrobial agents have clustered where the log(dose)max values are approaching zero. The natural agents appear as category 3 antimicrobials, exerting little antibacterial activity; however, they did prove to have a greater effect than the aqueous extractant solution. It is difficult to compare a variety of agents because a comparison needs to be made between molarity, weight and volume. The deviation from zero technique described earlier is one method with which to compare values with different units. This method allows the determination of the rate of antimicrobial action, which is very important in assessing a putative antimicrobial agent for inclusion in an oral hygiene product formulation. Antimicrobial agents that deviated significantly above zero, for example copolymer F127, had a log(dose)max 10 below the log(dose)max 15 indicating that the antimicrobial action required more than 10 min for a significant effect. The copolymer F127 shows some action against Strep. mutans, but this is incidental to its role as a surfactant in oral hygiene product formulations. Some plant extracts can replace synthetic antimicrobial agents so long as a more concentrated plant extract could be included in a formulation (assuming the natural antimicrobial agent conforms to other properties required of an antimicrobial agent as listed earlier, e.g. toxicity). Ethanol proved to be a ‘poor’ antimicrobial agent for inclusion in mouthrinses, but may be good in circumstances where dilution effects are not important. Ethanol is often included in mouthrinse formulations because it gives a taste ‘bite’ as well as dissolving other mouthrinse ingredients. The mode of action is thought to be by membrane disruption due to the hydrophilic hydroxyl group (Kubo et al. 1995).

The effect of an active agent must also be tested in vivo because homogeneous broth cultures do not grow and behave in the same way as an immobilized heterogeneous biofilm such as dental plaque. Wilson (1996) described the in vitro systems available to evaluate the susceptibility of oral bacteria to antimicrobial agents. Use of the constant depth film fermenter (University of Wales, Cardiff, UK) is one such method that can assess the activity of compounds against bacteria, under conditions similar to those that would exist in vivo, by enumerating bacteria on the substratum surface (Pratten et al. 1998). Flow chambers have also been used in conjunction with image analysis to study the effect of compounds on the formation of a biofilm (Morgan and Wilson 2000). Heterogeneity through the depth of the biofilm, exclusion of the biocide to the underlying layers, quenching of the biocide in the outer layers of the biofilm and induction/suppression of genes on attachment to surfaces may all affect antimicrobial susceptibility (Brown and Gilbert 1993). Micro-organisms in a biofilm are far more resistant to antimicrobial agents than when dispersed in a liquid medium (Carpentier and Cerf 1993), suggesting that in vitro tests are in fact more difficult to use to determine antimicrobial efficacy. Future efforts should determine the efficacy of antibacterial agents on non-planktonic cells. Indeed, the metabolism of Strep. mutans biofilm can be measured by a microcalorimeter (Morgan and Beezer 1998).

Microcalorimetry has proved to be a useful instrument in the screening of potential antibacterial agents against Strep. mutans by looking at the biological effect. This technique could be used by the pharmaceutical industry to help to select compounds and the concentration at which to use them prior to further microbiological analysis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography

Johnson and Johnson Ltd supported this work. The authors thank Sandra Sidney and Mike Harris for their intelligent discussions.

Bibliography

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. Bibliography