Inhibition of peptidase and giycosidase activities of Porphyromonas gingivalis, Bacteroides intermedius and Treponema denticola by plant extracts

K. A. Homer\ F. Manji^ and D. Beighton^ ^Hunterian Dental Research Unit, London Hospital Medical College, Turner Street, Whitechapel, London, El 2AD, UK: ^Oral Health Research Unit, Medical Research Council, KEMRI, PO Box 20752, Nairobi, Kenya.

Homer KA, Manji F and Beighton D: Inhibition of peptidase and giycosidase activities of Porphyromonas gingivalis, Bacteroides intermedius and Treponema denticola by plant extracts. J Clin Periodontol 1992; 19: 305-310. Abstract. Aqueous extracts from 5 plants used widely in Kenya as chewing sticks (mswaki) for the control of oral hygiene were tested for their ability to inhibit extracellular peptidase and giycosidase enzyme activities produced by the periodontopathic bacteria Porphyromonas gingivalis (formerly Bacteroides gingivalis), Bacteroides intermedius and Treponema denticola. The plants studied were Rims natalensis, Cupressus hisitanica, Sida cordifolia, Olea africana and Euclea divinorum. Protease activities, including glycylprolyl dipeptidase and trypsin-like activities of P. gingivalis, chymotrypsin-like and glycylprolyl dipeptidase activities of B. intermedius and the trypsin-like activity of T. denticola, were particularly affected by extracts from Rhus natalensis and Euclea divinorwn. Giycosidase activities were generally less affected with the notable exceptions ofthe inhibition of/?-mannosidase activity of P. gingivalis by all extracts and the inhibition of neuraminidase activity of T. denticola by Rhus natalensis and Euclea divinorwn. Generally, these same proteolytic and glycosidic activities were inhibited by tannic acid and to lesser extents by gallic acid and gallic acid methyl ester. An inhibitory component, present in all extracts, exhibited physical and chemical properties identical to those of tannic acid. The inhibition of these enzyme activities is likely to reduce the virulence of these periodontophathic bacteria and to reduce the rate of dental plaque formation.

In many countries, selected plants are fashioned into chewing sticks and used to maintain good oral health in order to reduce the incidence of dental diseases, especially dental caries and periodontitis. The mechanisms of action by which they might exert a clinical effect are not fully understood. However, we have recently shown (Homer et al. 1990) that aqueous extracts of plants used in Kenya as chewing sticks have marked effects on the degradation of model proteins and glycoproteins by putative periodontopathogens (Treponema denticola, Bacteroides intermedius and Porphyromonas {Bacteroides) gingivalis). The effects of plants extracts on bac-

terial degradative activities have not previously been studied, but various plant extracts have been shown to inhibit, in vitro, the glucosyltransferase activity and growth of Streptococcus mutans and Streptococcus sobrinus (Wolinsky & Sote 1983, 1984, Kohda et al. 1985, Southard et al. 1984, Dzink & Socransky 1985, Wu-Yuan et al. 1988, Gazi et al. 1987, Stalfors 1967) and growth of Bacteroides spp. (Rotimi & Mosadomi 1983, 1987, Akpata & Akinrimisie 1977). In this paper, we have determined the effects of plant extracts on extracellular degradative enzyme activities (peptidases and glycosidases) of subgingival

Key words: Bacteroides intermedius; giycosidase; peptidase; plant extracts; Porphyromonas (Bacteroides) gingivalis; tannic acid; Treponema denticola Accepted for publication 26 February 1991

plaque bacteria; inhibition of these bacterial peptidase and giycosidase activities is likely to reduce their virulence and growth in vivo. As Wu-Yuan et al. (1988) have previously demonstrated that inhibitory factors in Chinese nutgall include phenolic compounds (gallotannins) we have also determined the effects of authentic plant phenolic compounds on these bacterial enzyme activities. Material and Methods Growth of micro-organisms

The strains of P gingivalis, B. intermedius and T. denticola used in this study


Homer et al.

and their sources are shown in Table 1. P. gingivalis and B. intermedius strains were grown on Fastidious Anaerobe Agar (Lab M, Salford, Lanes, UK) at 37°C in an anaerobic chamber (Don Whitely, Shipley, W. York, UK) for 2 days. Bacterial colonies were suspended in 50 mmol/L N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid buffer, pH 7.5 (TES buffer), washed twice in TES buffer and stored at -20°C. Due to the difficulties of harvesting T. denticola from agar it was grown in peptone-yeast extract-serum medium (Smibert 1985), modified by the addition of only 1% (V/V) serum, for 3 days at 37°C. Cells were harvested by centrifugation, washed 2 x in TES buffer, and stored at -20°C. Source of plants and preparation of extracts

Aqueous extracts were prepared from five species of plant {Rhus natalensis, Cupressus hisitanica, Sida cordifolia, Olea africana and Euclea divinorum, referred to here as Extracts A, B, C, D and E respectively) which are the plants most frequently used by Kenyan schoolchildren as mswaki, or chewing sticks (Homer et al. 1990). Briefly, approximately 500 g of twigs from each species of plant were homogenized and allowed to stand in distilled water at 4°C for 24 h. The aqueous extracts were decanted, clarified by centrifugation and freeze dried prior to storage at 4°C. Extracts were reconstituted by dissolving each in TES buffer at a concentration of up to 2 mg of freeze-dried extract per ml of buffer. Protein determination

Proteins were extracted from bacteria by boiling for 5 min in 0.5M NaOH as described by Herbert et al. (1971) and quantified with the Coomassie Brilliant Blue dye-binding assay (Pierce, Rockford, Illinois, U.S.A.) using bovine sertim albumin (Sigma) as the standard. Assay of enzyme activity

Enzyme activities were determined by following the hydrolysis of fiuorogenic substrates, based upon previously described methods (Zimmerman et al. 1977). The substrates (Sigma, Poole, Dorset, UK) used are shown in Table 1; they were dissolved in a minimum volume of dimethylsulphoxide and di-

luted in TES buffer to give a final concentration of 20 [iglml. Stock solutions of plant extracts and authentic plant phenolic compounds, tannic acid, methyl ester of gallic acid and gallic acid (Sigma) were prepared in TES buffer and assays were set up in duplicate in microtitre trays with 50 [A of bacterial cell suspension and 50 fA of plant extract. These were pre-incubated at 37°C for 15 min to allow the plant extract to interact with bacterial enzymes and then 100 jA of the appropriate substrate solution was added. Plant extracts were tested in the assay system at final concentrations of 500, 250, 100, 50 and 10 /ig/ml. Rates of uninhibited reactions were detennined in the absence of plant extracts with 50 fA of TES buffer added to maintain the total volume at 200 ^1. Fluorescence was measured on a microtitre plate reading attachment fitted to a Perkin-Elmer LS-3B fiuorescence spectrometer using an excitation wavelength of 380 nm and emission wavelength of 460 nm. Control assays contained 50 fA TES buffer in

place of bacterial suspension to allow for the nonbiological breakdown of substrate.

Thin-iayer chromatography of plant extracts

Aqueous solutions of the plant extracts and authentic plant components were prepared to a final concentration of 2 g/1 and 5Q [A was applied to thin-layer silica gel plates (20x20 cm silica gel plates, layer thickness 0.2 mm, E. Merck, Darmstadt, W. Germany) in 1 fA aliquots. Plates were developed in tanks containing methanol:water (60:40 by volume). Silica gel was removed from 1 cm sections of each track into 200 fx\ of TES buffer, centrifuged to remove the silica gel and 50 fA of the supernatant was tested for its ability to inhibit the hydrolysis of a-benzoyl-arg-AMC by P. gingivalis strain LM9. Duplicate plates were run, viewed for spots under short wave ultra violet light and stained with 1% (W/V) ferric chlor-

Table I. Specific rates of hydrolysis of fluorescent substrates (nmol/hr/mg bacterial protein) and coefficient of variation (100 x Standard Deviation/mean) by strains of P. gingivalis, B. intermedius and T. denticola; each value represents the mean of at least 5 determinations LM9^



75.0(4.9)49.5(8.6) 111.5(2.2) 59.1(5.0) 95.8(6.1) 678.3(9.5) 208.3(1.7) 94.6(7.7)

68.6((2.2)^ 23.2(5.6) 67.8(3.8) 41.3(5.5) 165.9(10.6) 122.0(9.9) 76.5(2.0) 79.7(3.4)

83.3(2.7)6.7(21.6) 50.5(2.8) 26.8(3.5) 233.6(9.4) 162.8(7.1) 87.8(1.8) 82.0(7.3)

(b) B, intermedius




L-Arginine-AMC L-Leucine-AMC 4-MU-a-D-glucoside 4-MU-/9-D-glucorunide N-CBZ-Gly-gly-arg-AMC N-Succinyl-ala-ala-phe-AMC Gly-pro-AMC

50.3(6.3) 30.2(13.2) 70.2(3.4) 15.3(13.7) 8.2(11.0) 17.9(9.5) 54.5(5.8)

35.1(22.2) 61.7(4.7) 314.5(3.9) 43.5(26.3)

36.6(8.3) 18.4(10.3) 40.6(5.5) 16.8(7.1) 10.0(4.3) 10.8(12.5) 31.3(8.1)



60.0(6.9) 163.4(2.8) 142.6(3.9)

12.5(3.3) 33.0(2.5) 39.5(2.9)

Enzyme substrates' (a) P. gingivalis L-Arginine-AMC 4-MU-jff-D-mannopyranoside 4-MU-N-acetyl-y5-D-glucosaminide 4-MU-N-acetyl-y5-D-galactosaminide N-CBZ-Gly-gly-arg-AMC Na-benzoyl-L-arginine-AMC Gly-pro-AMC N-succinyl-ala-phe-lys-AMC

(c) T. denticola 2'-(4-MU)-a-D-N-acetyl-neuraminic acid L-proline-AMC Na-benzoyl-L-arginine-AMC



"'/! gingivalis strains LM9 and LM14 and B. intermedius strain N275/70A were provied by Dr. H. Shah, LHMC; "^P. gingivalis strain HG185 and B. intermedius strains HG659 and HG772 were provided by Professor J. de Graaff, ACTA, Amsterdam; ' T denticola strain ASLM was provide by Dr. W. J. Loesche, Ann Arbor, USA; "^T denticola strain ATCC 33520 was purchased from American Type Culture Collection. ND = not detected. 'AMC = 7-amido-4-methylcoumarin and 4-MU = 4-methylumbelliferyl. ^Mean (coefficient of variation).

Enzyme inhibition by plant extracts ide solution to reveal phenolic components including tannic acid.


Activity {% of control) 120

Analysis of results

All assays were performed in duplicate and from all values the fiuorescence of control assays set up with substrate alone were subtracted. For assays without plant extract the specific rates of hydrolysis of each substrate were calculated and are given as nmol substrate hydrolyzed/hr/mg protein. Coefficients of variation for rates of hydrolysis of substrates by test organisms were calculated. The fiuorescence obtained in assays with plant extracts or authentic compounds was expressed as a % of the control value without plant extract. The concentration of plant extract or authentic compound required to caused 50% inhibition of the hydrolysis of each substrate was determined.

Results Rates of hydrolysis of substrates

~"*~ Extract A ~t~ Extract B -^

Extract C

- H - Extract D Extract E

100 200 300 400 500 Concentration of piant extract (ug/mi)


Eig. I. Inhibition of L-arginine arylamidase activity in P. gingivalis by plant extracts. Extract A, B, C, D and E represent extracts of plant species as species as described in materials methods.

Activity {% of control) 120

The specific rates at which each of the organisms hydrolyzed the substrates, in the absence of plant extract, are shown in Table 1. The coefficients of variation for rates of hydrolysis of substrates were generally less than 10%. Similar low variation was found in all assays with plant extracts included.

Extract A Extract B Extract C Extract D Extract E

Effects of plant extracts and authentic compounds on substrate hydrolysis

The inhibition of the P. gingivalis arginine arylamidase activity, as determined by the relative rates of hydrolysis of the L-arginine-AMC substrate, by the five plant extracts over a range of concentrations is shown in Fig. 1. Extracts A and E were the most effective inhibitors of this enzyme, reducing the activity to less than 30% of the control value at a concentration of 10 /ig/ml. In contrast, the rate of degradation of the giycosidase substrate, 4-MU-N-acetylglucosamine, was relatively unaffected in the presence of any of the 5 extracts unless concentrations of >250 or >500 fig/m\ were used (Fig. 2). The concentration of each plant extract and of each authentic plant compound required to inhibit, by 50%, the hydrolysis of each substrate by the three subgingival plaque bacterial species are shown in Table 2.

100 200 300 400 500 Concentration of piant extract (ug/ml)


Eig. 2. Inhibition of N-acetylglucosaminidase activity in P. gingivalis by plant extracts. Extracts A, B, C, D and E represent extracts of plant species as described in materials and methods.

Partial characterization of inhibitory activity

Each plant extract contained components that reacted with FeCl3 and fiuoresced with UV light suggesting the presence of a phenolic compound. A component, inhibitory for the hydrolysis of a-benzoyl-arg-AMC by P. gingivalis, was present in each extract and co-eluted with tannic acid on silica gel TLC plates.


P. gingivalis, B. intermedius and T. denticola, often in conjuction with other subgingival plaque bacteria, are associated with periodontitis (Loesche et al. 1985, Slots & Listgarten 1988, Haffajee et al. 1988). Many virulence mechanisms have been attributed to these organisms, including the production of extracellular enzymes (proteases and glycosidases) capable of degrading host-

m 308

Homer et al.

derived serum and tissue components (Fine & Mandell 1986, Van Steenbergen et al. 1987, Van Winkelhoff et al. 1988). The enzymes studied here include putative virulence determinants and activities necessary for the degradation of host-derived glycoproteins to yield metabolizable substrates to support bacterial growth in vivo; such functions are, of course, not mutually exclusive. Virulence determinants studied in detail by others include the trypsin-like enzymes of P. gingivalis which degrade basement collagen type IV and denatured collagen (Sorsa et al. 1987, Suido et al. 1987), bovine serum albumin and ovalbumin (Yoshimura et al. 1984, Tsutsui et al. 1987) and glycylprolyl dipeptidase that degrades partially denatured collagen type I (Abiko et al. 1985) and a prolinerich protein from saliva (Grenier & McBride 1987). Trypsin-like activity is produced by T. denticola but it is suggested that its function in vivo is to cleave N-terminal arginine residues rather than possessing a true endopeptidase activity (Ohta et al. 1986). The role of giycosidase activities in the virulence of these organisms is not clear, though the action of ^-mannosidase, neuraminidase and y9-N-acetylhexosaminidase might remove oligosaccharide sidechains from glycoproteins rendering the protein core more susceptible to proteolysis (Sharon & Lis 1982,

Gottschalk & Fazekas de st. Groth 1960). The hberated carbohydrate and amino sugar moieties could then serve as nutrient sources for many dental plaque bacteria. All of these enzyme activities would act sequentially and collectively to degrade glycoproteins either by removing N-terminal amino acids (arylamidase activity), by removing individual sugars from side-chains (glycosidases) and by cleaving peptide bonds within proteins to facilitate further enzymatic degradation (endopeptidases). Inhibition of any of these activities might therefore reduce the virulence of these bacteria and reduce their ability to obtain nutrients for growth (Beighton et al. 1986, 1988, Smith & Beighton 1986, 1987, Ter Steeg et al. 1988). There is evidence, primarily from in vitro studies, that plant extracts could exert a profound effect on the composition and accumulation of bacterial deposits on the dentition and so infiuence the onset of disease. Extracts from other plants inhibit many physiological properties of mutans streptococci (Wolinsky & Sote 1983, 1984, Kohda et al. 1985, Southard et al. 1984, Dzink & Socransky 1985, Wu-Yuan et al. 1988) and may be bactericidal for a number of oral bacteria, including Bacteroides spp. (Akpata & Akinrinisie 1977, Rotimi & Mosadomi 1987). In this study we have extended these

observations and have demonstrated that extracts from plants commonly used as chewing sticks in Kenya exert inhibitory effects on potential virulence factors of three putative periodontopathic bacteria. It is also likely that these extracts would inhibit similar classes of enzymes produced by other dental plaque bacteria and inhibit the formation of both supra- and sub-gingival plaque. In a previous study (Homer et al. 1990), we also demonstrated that extracts A and E inhibited protease activity of these three bacterial species to a greater extent than the other extracts, and here we have found that, overall, plant extracts A and E also had greatest effects of the rates of hydrolysis of the synthetic protease and giycosidase substrates. Preliminary investigations revealed that the inhibitory activity of the extracts co-eluted with tannic acid on TLC silica gel plates. Although more comprehensive study of the structure of the inhibitory components is required, tannic acid and its derivatives are common metabolic end products in plants. Tannins have already been implicated as the active factors in Chinese nut gall {Melaphis chinensis) extracts which inhibit mutans streptococci growth, glucan synthesis and aggregation (WuYuan et al. 1988). Furthermore studies

Table 2. Concentrations of plant extracts and phenolic compounds (//g/ml) required to give 50% inhibition of enzyme activity in P. gingivalis, B. intermedius and T. denticola. Plant extract Enzyme substrates (a) P gingivalis L-arginine-AMC 4-MU-/i-D-mannopyranoside 4-MU-N-acetyl-^-D-glucusaminide 4-MU-N-acetyl-/y-D-galactosaminide N-CBZ-Gly-gly-arg-AMC Na-benzoyl-L-arginine-AMC Gly-pro-AMC N-succinyl-ala-phe-lys-AMC (b) B, intermedius \ L-arginine-AMC L-leucine-AMC 4-MU-a-D-glucoside 4-MU-/?-D-glucoronide N-CBZ-gly-gly-arg-AMC N-succinyl-ala-ala-phe-AMC Gly-pro-AMC (c) T denticola 2'-(4-MU)-a-D-N-acetylneuraminic acid L-proline-AMC Na-benzoyl-L-arginine-AMC









0.05-0.1 0.5-1.0 50-75 25-50 25-50 25-50 75-100 0.5-1.0

0.1-0.5 5-10 75-100 50-75 5-10 5-10 50-75 0.5-1.0

0.5-1.0 5-10 75-100 75-100 25-50 25-50 75-100 0.5-1.0

250 >250 100-250 >250 100-250

50-100 50-100 >500 >500 >500 >500 >500 >250

10-50 50-100 >250 >500 >500 >500 >250 >500

10-50 >250 >500 >500 100-250 50-100 >250 10-50

50-1 00 >500 >500 >250 >250 >250 10-50

25-50 5-10 50-75 5-10 5-10

75-100 50-75 75-100 5-10 25-50 25-50 75-100

>500 >250 >500 100-250 >500 >500

>250 50-100 >500 100-250 50-100 50-100 >250

>250 >250 >500


5-10 50-75 75-100 50-75 25-50 25-50 50-75

100-250 >250 >250

50-100 50-100 >500 100-250 50-100 50-100 100-250

0.1-0.5 75-100 5-10

10-25 25-50 25-50

25-50 75-100 75-100

100-250 >500 >500

>500 >250 >500

>500 100-250 >250

10-50 >250 100-250


Inhibition of peptidase and glycosidase activities of Porphyromonas gingivalis, Bacteroides intermedius and Treponema denticola by plant extracts.

Aqueous extracts from 5 plants used widely in Kenya as chewing sticks (mswaki) for the control of oral hygiene were tested for their ability to inhibi...
10MB Sizes 0 Downloads 0 Views