Antibacterial activity and mode of action of plant flavonoids against Proteus vulgaris and Staphylococcus aureus

Antibacterial activity and mode of action of plant flavonoids against Proteus vulgaris and Staphylococcus aureus

Phytochnnictry. Vol. 26. No. 8. pp. 22314234, Printed in Great Britain. 1987. 0031-9422187 ~3.00 + 0.00 Q 1987PcrgamosJoutnslsLtd. ANTIBACTERIAL A...

359KB Sizes 2 Downloads 86 Views

Phytochnnictry. Vol. 26. No. 8. pp. 22314234, Printed in Great Britain.

1987.

0031-9422187 ~3.00 + 0.00

Q 1987PcrgamosJoutnslsLtd.

ANTIBACTERIAL ACTIVITY AND MODE OF ACTION OF PLANT FLAVONOIDS AGAINST PROTEUS VULGARIS AND STAPHYLOCOCCUS AUREUS AKIHISA MORL CHIKAO NISHINO~, NOBUYASU ENOKI and SHINKICHI TAWATA* Mitsubishi-Kasei Institute of Life Sciences. I 1 hlinamiooya, Machida-shi. Tokyo 194. Japan; ‘Department ol’ Agricultural Chemistry. University of the Ryukyus. Nishiharacho. Okinawa 903-01. Japan (Revised received 16

Key Word Index-Uaeognus synthesis; RNA synthesis.

g/&o;

January 1987)

Elacgnaceac; ( - )-epigallocatechin; ftavonoids; antibacterial activity; DNA

Abstract-( -)-Epigallocatechin. an antibiotic found in EIaeagnus g/&a, and 28 other related plant flavonoids were tested for their antibacterial activity against Proteus uulgoris and Staphylococcus aureus. A free 3’,4’,5’-trihydroxy B-ring and a free 3-OH were necessary for antibacterial activity. DNA synthesis was predominantly inhibited by the active flavonoids in P. vulgaris, whereas RNA synthesis was inhibited in S. oureus.

Incorporation of radioactive precursors

INTRODUCTION (-)-Epigallocatechin (compound 27 in Table 1) in the bark of an Okinawan medicinal plant, Eheagnus glabra Thunb., (Elaegnaceae) has been found to be an antibiotic against the human skin bacterium, Staphylococcus epidermidis [I]. This flavonoid has also shown lo have antibacterial activity against Proteus vulgaris (Gramnegative) and S. aureus (Gram-positive), and is cytotoxic lo HeLa cells [I]. (-)-Epigallocatechin (27) belongs to the flavonoid class. Antibacterial and antifungal activities of flavonoids have been reported [Z-6]. but no systematic study on antibacterial activity of Ravonoids has yet be-en carried out. In this work, 27 and 28 related flavonoids are assayed for antibacterial activity against P. vulgaris and S. atireus. This paper also reports structure-activity relationship and effect of several active flavonoids on macromolecular and lipid synthesis in the bacteria.

RESULTS

Antibacterial activity

Proteus vulgaris. The incorporation experiments were performed with 14-16 and 27. Incorporated radioactivity of the precursors was measured depending upon Bavonoid concentration and culture period. Radioactivity at each concentration of 14 showed comparable counts lo those of controls at any culture period in all the precursors. This means that neither macromolecular (DNA, RNA and protein) nor lipid synthesiswere inhibited by 14. Radioactivity diminished according to increasing flavonoid concentration in 1 hr culture of IS, 16 and 27, the degree (%) of inhibition of macromolecular and lipid synthesis in 1 hr culture is shown in Fig. I (A). DNA synthesiswas most strongly inhibited, followed by RNA synthesis. Staphylococcus aureus. 7,8-Dihydroxyflavone (2) was used in addition lo IS, 16 and 27. Similarly, yOinhibition of the flavonoids lo the syntheses is plotted against davonoid concentrations on the basis of the data at 1 hr culture, and shown in Fig. l(B). RNA synthesis was predominantly inhibited by 2,16,and 27 (Fig. 1A,a,cand d) and DNA synthesisby IS (Fig. 1 B. b). Compounds 15 and 27 exhibited much weaker inhibitory effect on the synthesesin S. aureus than that in P. oulgaris.

Antibacterial activity of the flavonoids is displayed in Table 1. At SO-200/.4ml (MIC) (lOO&ml mostly), six

fiavonoids were active against P. uulgaris, whereas five were active against S. aureus. Among the active flavonoids. datiscctin (8) and quercctagetin (14) were active against P. vulgaris only (100 &ml) and 7,8dihydrotlavone (2) against S. aureus only (100 &ml). Robinetin (IS), myricetin (16), (+)-dihydrorobinetin (24) and (-b epigallocatechin (27) were active against both bacteria (50-200 &ml).

tAuthor 10 whom enquiries shouldbe addressed.

DISCUSSION

Although antibacterial activity has often been described for flavonoids [2-S], structure-activity relationships are not knownas theyarein thecaseofantifungalactivity [6]. No specific structural factor was effective against both P. vulgaris or S. aureus from the present data. However, when the structures of the flavonoids (15.16.24 and 27), which were active against both bacteria, were inspected, several important structural factors are apparent. A free hydroxyl group on the aromatic A- and B-ring was necessary.The flavonoids without such free hydroxyls (1 and 17) possessedno activity, whik all the active species had them. Furthermore, if the hydroxyls of 27 were masked (28 and 29), activity was lost.

2231

2232

A. MORI et al. Table 1. Structures and antibacterial activity (WC) of Itsvonoids against Prorw

uulguris and Srophylococcus

Substituent on aromatic rings 5

Flavonoid

Flavones

Flavone (1)

‘I,&Dihydroxyfiavone (2) 6.7-~hydroxy~vone (3) Chrysin (4) Luteolin (5) Apigcnin-‘IA’-dimethyl ether (6) Flavonols Golangin (7) Datisctin (8) Kaempkrol (9) F&tin (10) Quercetin ( 11) Rhamnetin (12) Morin (13) Quercctagetin (14) Robinetin (15) Myricetin (16) Flavanoncs* Fkvanonc (17) 4’.5.7-Trihydroxytlavanone (18) Eriodictyol (19) Homocridictyol (20) Hespertia (21) Fkvanonols t ( - )-Fustia (22) (+)-Taxifolin (23) (+)-Dihydrorobinetin (24) CUcchinst (+ FCatcchin (25) ( - )-Epicatcchin (26) (-)-Epigallocatcchin (27) Methyl ether of 27 (28): Acetate of 27 (29)$

67

82’3’4

H

H

H

H

HOHOHHH

H OH OH

OHOHHHH H OH H H OH H

OH OH OH OH H OH OH OH OH H OH H

H H H H H H H H OH H H H

OCH,H H OH H H OH H OH OH H H OH H H OH H H OCH,H H OH H OH OH H H OH H H OH H H H HHH

MIC (r%mlG 5

P. vulgaris

S. aweus

H

H

-

-

H

H

-

loo

H H OH

H H H

-

-

H H H H OH OH OH H OH OH OH

OCH, H H OH OH OH OH OH 0 OH OH H

H H H H H H H H H OH OH H

100 100 100 100 50 -

100 100 -

HHH

H H

~U~CUS

H OH

OH

H

OH

H

H

H

OH

H

-

-

OH

H

OH

H

H

OH

OH

H

-

-

OH OH H OH

H H H H

OH OH OH OH

H H H H

H H H H

OCH,OH H OH OCH,H OH OH H (2& h) OH OH H (Za.3fi)

-

-

-

-

H OH OH

H H H

OH OH OH

H H H

H H H

OH OH OH

OH OH OH

OH (2.x. 38) H (2a.36) H (2a. 3a)

200 -

200 -

OH

H

OH

H

H

OH

OH

OH (2a, 3a)

50

100

OCH,OCH,OCH, (Za, 3a-OH) OAc OAc OAc (2a 3a-OAc)

-

-

OCH, H

OCH, H

H

OAc H

OAc H

H

*AI1 arc racemates. tA8 are optically active compounds. $3-OH is fret in 28, whcras it is acetylared in 29. §- means inactive at 200 p%rnl in MIC.

The 3’,4’,5’-trihydroxy B-ring and the 3-OH were common structural moieties to all the active flavonoids. This type of B-ring coupled with the 3-OH may be, accordingly, the most important structural unit for antibacterial activity against P. ulrlgaris and S. aureus. This was confirmed by the data for the following two pairs of flavonoid series: inactive 10 and 26 were changed to active 15 and 27, respectively, by adding a S’-hydroxyl in the Bring. As with the antifungal activity of tlavonoids [6], antibacterial activity of 24 and 27 is given by a non-planar flavonoid structure. This contrasts with the data of ref. 13) which illustrated the importance of the 2.3double bond (namely, molecular planarity) of Aavonoids for the activity. The importance of planarity of structure has also been reported in several pharma~lo~~i tests 17-93. As

found previously [ 1J, 2,3&s substituion (27) provides greater activity in the non-planar flavonoids series than 2,3-trots substitution (24). The 4-carbonyl group was less important, because of the activity of 27. Eli~~tion of the carbonyl group usually decreases biological activity of fiavonoids E9, IO]. However in the case of compound (27) which lacks a 4carbonyl group, the 3-OH might function in controlling the hydrophobicity of the molecule. It is known that lipid content is greater in the cell wall of Gram-negative bacteria than Gram-positive bacteria. More lipophilic 2 may be trapped into the cell wall of P. vulgaris, whereas it passes through the S. aureus cell wall; this may explain why 2 is active against S. aureus only. f -)-Epigallocatechin (27). which is hydrophilic, appears to permeate to P. vufgoris and S. ~lureus cell walls

Antibacterial

activity of tlavonoids

2233

A. Pmfws ntl~ D

c. t-1 - Evqlbc.&ch

b Yyw4hdl8)

Rcker.fln(l8)

(21)

lco

80

s60 :g

5

40

20

i

k 0

25 50 75 100 Concentmtion of fl0v0twid(~/ml)

0. Si~h~loconvs

WlM

o. 73 - Myhxy

d (-)[email protected])

- fbmftt)

r

I00

60

e

60

60

60

.-s f z C

‘ii

40

40

20

20

0

‘I”

25

50

75

100

0:

0

/

Y.3

75

100

Concentration of flavonoid (rg/ml)

Fig. 1. Inhibitory eNectof

Aavonoids on macromolecular

[DNA

(0).

RNA

(0)

and protein (A)] and

lipid (0)

synthesis in Proteus vulgaris (A) and Srophylucoccus aureus (B). The bacteria were treated with the indicated concentrationsof

the tlavonoids [ZS, SOand LOO&ml

for 2 (Ba). 14.15 (Aaand

B b)and 16 (A band Bc).and

200 and 400 pg/ml for 27 (A c and B d)] together with the radioactive precursors (0.1 mCi/ml) Inhibition

percent was calculated on the basis of the incorporation showed no inhibitory

data at

I

100.

shown in the figure.

hr culture. Since quercctagetin

(14)

action to synthesis in P. oulgmis. it is omickd.

at different rates. This is reflected in the discrepancy in inhibitory effect on the syntheses between the bacteria (compare Fig. I A, c with Fig. 1 B. d). Similar reasoning would explain the different effects of 15(Fig. 1 A, a and B, b). Hydrophobicity of 16 is weaker than that of 15, since there is intramolecular hydrogen bonding between the 4carbonyl group and 5-OH. As a result, the hydrophobicity of 16 allows it to pass through the cell walls of both P. oulgarb and S. aureus. With tlavonoids, their interaction with the cell surface of microorganisms seems to be important [63. With regard to the mode of action of the antibacterial flavonoids. it is generally recognized that DNA synthesis is most strongly inhibited by flavonoids in P. w/gori.s, while RNA synthesis is most effected in S. aureus. Flavonoids (e.g. 11) inhibit the synthesis of DNA, RNA and related macromolecules [9-12. 13 and 143 in other

pharmacological experiments. Intercalation and mispairing activity towards the nucleic acid bases [9, lo], have sometimes been proposed because of the similar planar structures of flavonoids and the basesof DNA and RNA. However, in this study, the non-planar llavonoids. 24 and 27, showed antibacterial activity of the 3’,4’,5’-trihydroxy B-ring coupled with the 3-OH must be important for activity. The B-ring, accordingly, may play a role in intercalation or hydrogen bonding with the stacking of nucleic acid bases, which is reflected in the inhibitory action on DNA or RNA synthesis. EXPERIMENTAL F/aoonoids. Catechins. M and 27 were isolated from E. g/&a [I], and 28 and 29 were prepared from 27. The other flavonoids were obtained commercially.

The fiavanones

(I 7-21)

were race-

2234

A. MORNez al.

mates, while the flavanonols (22-24) and catechins (25-29) were optically active (1 J. Radioactive precursors. [Z-“C’jThymidine (51 mCi/mM), [2“C]uridine (51 mCi/mM)and t.-[U-t*C]kucinc (348 mCi/mM) were purchased from Radiochemical Centre, Amersham Buckinghamshire, and [I-“C]aartic acid sodium salt (56 mCi/mM) was obtained from New England Nuclear Corp. 4ntibacrerial ucrioiry. The activity was measured by the standard agar dilution method for MIC dete~Mtion [is] with a heart infusion agar (E-MC10, Eiien Chemical Co.). Inocula of l/l00 dilution of Proteus vulgaris OX-19 or Staphylococcus aureus FDA 209 PJC-1, prepared by dilution of a fresh 18-20 hr (37”) broth culture. were applied to agar plates [Petri dish (9 cm in diameter)] each of which had been mixed with a concentration of a Ravonoid ([email protected] ml of S% DMSO). After incub ation ofthcagar phtcs for 18-20 hrat 37”. the minimal inhibitory concentration (MIC, &ml) was determined. Incorporation of radioacrive precursors into bucrerial macromolecules and Lipid.For the incorporation experiments, flavonoids, 2, 1416 and 27 were used 2 was active against S. aureus only, whereas f4 against P. vulgaris only. The others were effective against both bacteria. A bacterium (P. vulgaris or S. aureus) was grown in a heart infusion broth (E-MC 68, Eikcn Chemical Co.) overnight at 37”, after which the culture was diluted to IO times in volume with the broth and incubated to show 0.2 absorbancc at 600nm. This exponential stageof the bacterium was used for the incorporation experiments (5 ml culture for DNA, RNA and protein synthesis and 7 ml for lipid synthesis). Each radioactive precursor (0.1 mCi/l ml culture) and flavonoid (a specified quantity/2GQpl DMSO) were added to the cuhurc, and incubated. For m~urement of DNA, RNA and protein synthesis,samples (0.5 ml) were harvestedperiodically (at 1.1.5 and 2 hr) and the reaction was stopped by adding 5 “/,-TCA (2.5 ml). The TCA-insoluble residue wascollected on a glassfilter (Whatman GF/C). The filter was washed with TCA and EtOH. dried and counted in a tolucnc based scintillation liquid (IO ml).

For Lipid synthesis, a culture sample (1.0 ml) was periodically filtered through theglass filter which was subsequently extracted with CHCl,-MeOH-Hz0 (1:2:0.8) (2 ml). After addition of Ha0 and CHCI, (1 ml each), the soln was centrifuged (3000 rpm, IO min)toscpara~etheCHCI, layer. Radioactivityofapart (1 ml) of the layer was counted in the scintillation liquid (10 ml).

REFERENCES 1. Nishino, C., Enoki, N., Tawata, S, Mori, A., Kobayashi, K. and Fukushima, M. (1987) Agric. Blot. Chem. 51, 139. 2. Das, D. N. (1962) 1. Indian Chom. Sot. 39,849. 3. Ramaswamy, A. S., Jayaraman, S., Sirsi. M and Rao, K. H. (1972) Indian J. Exp. Biol. 10, 72. 4. Kubo. hi.. Kimuta, Y.. Gdani. T., Tani, T. and Namba, K. (1981) Planta Med. 43, 194. 5. Gnanamanickam, S. S. and Mansfield, J. W. (1981) Phyrochemistry 20, 997. 6. D’Neill, T. M. and Mansfield, J. W. (1982) Trans. Br. My&. Sot. 73,229. 7. Bjcldancs, L. F. and Chang, G. W. (1977) Science 197,577. 8. Hardigrce, A. A. and Eplcr, J. L. (1978) Murat. Res. 58.231. 9. Fcrrell. J. E., Chang-Sing, P. D. G., Locw, G., King, R., Mansour. J. M. and Mansour. T. E. (1979) Molec. Pharmucol. 16, 556. 10. Beretr. A.. Anton, R. and Stockt, J. C. (1978) Experienria 34, 1054. 11. Mchs.M.L.andMacGrcgor.J.T.(1981)Mura1.Res.88,317. 12. Ishitsuka, H., Ohsawa, C., 0hiwa.T.. Umcda, I. and Suhara, Y. (1982) Anrimicrob. Agenrs ChemorLr. 22.61 I. 13. Cody. V., Middleton. E. and Harborne, J. B. (cds) (1986) Plant flavonoids in biology and medicine. in Progress in Clinical and ~io~~ica~ Research, Vol. 213. Alan R. Liss, New York. 14. Havsteen, B. (1983) Biochem. Pharmacol. 32. 1141. 15. The JapaneseSocietyofChemotherapy (197S)Chemotherapy 23. I.