REVIEW ARTICLE NUMBER 63 ANTIMICROBIAL PROPERTIES OF …download.xuebalib.com/xuebalib.com.40571.pdf · Proanthocyanidins of Machaerium,floribundum (M, 1150) were better inhibitors

Phyfochemisrry, Vol. 30, No. 12. pp. 3875-3883, 1991 Printed in Great Britain.

003 I 9422/9 1 $3.00 + 0.00 C 1991 Pergamon Press plc

REVIEW ARTICLE NUMBER 63

ANTIMICROBIAL PROPERTIES OF TANNINS

AUGUSTIN SCALBERT

Lahoratoire de Chimie Biologique (INRA), Centre de Biotechnologie Agro-Industrielle, Institut National Agronomique

Paris-Grignon, 78850 Thiverval-Grignon. France

(Receiwd 14 June 1991)

Key Word Index-Vegetable tannins; polyphenols; plant defences; antimicrobial properties; iron deprivation; detoxification; biodegradation; tannase.

Abstract-Tannin toxicity for fungi, bacteria and yeasts is reviewed and compared to toxicity of related lower molecular weight phenols. The dependence of toxicity on tannin structure is examined. The different mechanisms proposed so far to explain tannin antimicrobial activity include inhibition of extracellular microbial enzymes, deprivation of the substrates required for microbial growth or direct action on microbial metabolism through inhibition of oxidative phosphorylation. A further mechanism involving iron deprivation is proposed. Many microorganisms can overcome plant defences based on tannins. They may detoxify tannins through synthesis of tannin-complexing polymers, oxidation, tannin biodegradation or synthesis of siderophores.

INTRODUCTION

Tannins are water soluble polyphenols which differ from most other natural phenolic compounds in their ability to precipitate proteins such as gelatin from solution [l]. This property (sometimes called astringency) is the reason for their past and present use in the tanning of animal skins. Tannins are commonly found in a large array of higher plant species of both herbaceous and woody types. They can accumulate in large amounts (often more than 10% of the dry weight) in particular organs or tissues which can be almost any plant part: bark, wood, leaves, fruits or roots [2].

Tannins are distributed in two groups according to their structures [2] (Fig. 1): proanthocyanidins (conden- sed tannins), e.g. 1 and 2, and hydrolysable tannins. Hydrolysable tannins are esters of phenolic acids and a polyol which is usually glucose. The phenolic acids are either gallic acid in gallotannins, e.g. 3, or other phenolic acids deriving from the oxidation of gallic acid in ellagi- tannins, e.g. 4. Tannin M,s vary widely (usually between 500 and 3000) according to the number of flavanol units in proanthocyanidins, the number of galloyl residues in galloyl esters or the eventual dimerization or polymeriz- ation of ellagitannins. Commercial tannins, used in most of the studies described in this review, are mixtures of either proanthocyanidins (wattle, quebracho), gallotan- nins (tannic acid) or ellagitannins (chestnut, oak, myrobalan). Proanthocyanidins are also found in many food products such as tea, cocoa, sorghum or carob pods.

It is beyond doubt that tannin-containing plants have had a significant evolutionary advantage over their enemies. Tannins may deter herbivores from predation [3]. They may also deter microorganisms, either by increasing resistance against pathogens [4, 51 or by protecting essential tissues such as wood against decay

[6]. Man has not ignored these remarkable properties of tannin-rich materials. Many of the timbers selected for their high durability, e.g. European oak and chestnut, black locust, and some eucalypts, are rich in tannins [7]. Increasing attention is also being paid to the use of tannins as antimicrobial agents [S], for example in wood preservation [9, lo] or prevention of dental caries [l l].

In this paper, the data on tannin toxicity against fungi, bacteria and yeasts are reviewed in order to examine the following questions: (i) is the toxicity of tannins different from that of simpler related phenols such as gallic acid or catechin? (ii) Is it possible to establish a relationship between the molecular structure of tannins and their toxicity? (iii) Is astringency the essential property in- volved in tannin toxicity or do alternative mechanisms exist? (iv) What are the different strategies used by micro- organisms to overcome the plant defences based on tannins. (v) What is known about tannin biodegrada- tion?

TOXICITY OF TANNINS

Toxicity of tannins towards microorganisms is well documented (Table 1). Toxicity studies have involved various fields of research: food science, wood science, soil science, plant pathology, pharmacology and human and animal nutrition.

Toxicity is usually estimated by measurement of the reduction of the in vitro growth of mycelium for filamen- tous fungi, and by plate count methods, disk diffusion methods, nephelometry or respirometry for bacteria and yeasts. Some biochemical parameters, characteristic of the metabolism of certain microorganisms, are also used including cellulose degradation, glucan synthesis, nitrate, methane or ethanol production.

3875

3876 A. &ALBERT

HO.

bH

R

I H 2 OH

SH 7

60H Y

9H Ii

10 H c

It OH H

12 OH G

G-G = ai

6

H

Fig. 1. Chemical structures of proanthocyanidins, hydrolysable

tannins and other structurally related phenols. 1, procyanidin trimer, 2, prodelphiniclin trimer; 3, heptagalloylglucose; 4, pedunculagin; 5, catechol; 6. pyrogallol; 7, gallic acid; 8,

(+)-catechin; 9, (-hepicatechin; 10, (- )-epicatechin gallate; Il. (- )-epigallocatechin; 12, (- )-epigalIocatechin gallate; 13, ellagic

acid.

Tannins were thus shown to inhibit the growth of many filamentous fungi. The minimum inhibitory concentra- tion (MIC) is usually higher than 0.5 g 1-l (Fames annosus [17]) and often reaches 10-20 g 1 -I (Merulius lucrymans [18]). Germination of Crinipellis perniciosa spores is inhibited at a tannin level of 0.25 g l- ’ [5]. Yeasts seem to be more resistant. Some species are inhibited at a tannin level of 25 g I - ‘, whereas others require levels as high as 125 g I - ’ 123 J. The MIC for bacteria is usually lower and can vary between 0.012 and I gl-’ [28. 373.

Tannins and lower molecular weight phenols

Tannins are ligands which may form soluble or in- soluble complexes with many polymers such as proteins and polysaccharides [42]. They differ in that respect from most simpler phenols such as catechol (5), pyrogallol (6), gallic acid (7), catechin (8) and other flavanols [43]. In order to ascertain if tannin toxicity is due to these characteristic properties, their toxicity has often been compared to that of related phenols. At equivalent weight concentration, gallic acid and catechin were shown to be as toxic as wattle or myrobalan tannins for Chaetomium cupreum [13], and catechol as toxic as a tannin (of unknown origin) for Fames annosus [17]. On the other hand, pyrogallol and gallic acid were shown to be much less toxic than tannic acid for methanogenic bacteria [40]. Similarly, the MIC of catechol and gallic acid for Cellvibrio jiilvus and Bacillus s&i/is is four to 40 times higher than that of tannic acid or carob pod extract (condensed tannin) [373; however, the same study did not show significant differences between MIC of these same polyphenols for other bacteria such as Shigellu dys- enreriae or Streptococcus cremoris. Four purified dimeric theaflavins were shown to be as inhibitory as the related (-)-epicatechin gallate (10) or (-)-epigallocatechin gall- ate (12) to Cfostridium botulinurn, but significantly more inhibitory than the related ungalloylated flavanols (9) and (11) [24]. Proanthocyanidins of Machaerium,floribundum

(M, 1150) were better inhibitors of Pseudomonas multo- philia and Enterobacter cloacae than (+)-catechin (8) or (-)-epicatechin (9) [26]. It thus appears that if tannins are more toxic that related phenols to some species, this difference cannot be generalized to all microorganisms.

Tannin structure and toxicit)

The effect of tannin molecular structure on toxicity is largely unknown. Most of the authors have only com- pared the toxicity of raw tannin extracts, tannic acid but also other hydrolysable tannins such as of myrobalan, oak, chestnut and condensed tannins from wattle, que- bracho, sorghum, cocoa or carob pod. Hydrolysable and condensed tannins have sometimes been compared: their toxicities toward filamentous fungi [I33 or bacteria [35,37] were usually not found to differ significantly. However, yeasts were more sensitive to chestnut tannin than to quebracho tannin, and even more to tannic acid [23]. Very few studies have been carried out with purified tannin of known molecular structures [24, 283. More such studies are needed before drawing further conclu- sions.

Two structural features have, however, been discussed in relation to toxicity toward microorganisms; the

Antimicrobial properties of tannins

Table I. Microorganisms susceptible to the toxic effects of tannins

Microorganism Tannin substrate Ref.

3877

Filamentous fungi Aspergillus niger Botrytis cinereo Chaetomium cupreum Colletotrichum graminicola Coniophora olivacea Coriolus versicolor

Crinipellis perniciosa Fames annosus Gloeophyllum trabeum

Meru1iu.s lacrymans Penicillium sp. Poria monticola Trametes hirsuta Trichoderma viride Various preharvest seed moulds

Yeasts Saccharomyces cerevisiae Various yeast species

Bacteria Bacillus subtilis Bacillus stearothermophilus Clostridium botulinurn Desuljotomaculum nigrijicans Desulphovibrio Enterobacter cloacae Pseudomonas maltophilia Nitrobacter Nitrosomonas Proteus vulgare Staphylococcus aureus Photobacterium phosphoreum Pseudomonas solanacearum Polyangium Sporocytophaga Streptococcus mutans Streptococcus mutans Streptococcus mutans Streptococcus sobrinus Various soil bacteria Soil inoculum Various bacteria Food born pathogenic bacteria Various bacteria causing

diarrhoeal diseases Methanogenic bacteria Rumen fluid

M, W Strawberry CT M, W CT Eucalypt T Ironwood ET White oak ET Cocoa extract T Ironwood ET White oak ET TA TA White oak ET TA CT Sorghum grain

Sorghum grain TA, Ch, Q

Flavanols and purified theatlavins TA

Cl21 c43 Cl31 1141 Cl51 Cl61 16, 161 PI Cl71

Cl61 WI Cl81 161 Cl91 1201 c211

WI 1231

c241 c251

Machaeriumj7oribundum and cotton CT [26]

Cl-, HT, TA

Purified HT TA M, W

Leaf extracts PGG and purified oligomeric CT Purified HT

Flavanols and flavanols gallate Ch, W CT Carob pod extract, TA Flavanols and purified theatlavins

Tea extracts TA Lotus corniculatus leaves

c27l

C281 ~291 WI

c311 ~321 c331

c341 c351 C361 1373 C381

c391 C403 c411

Abbreviations: Ch: chestnut tannin; CT: condensed tannin; ET: ellagitannin; GT: gallotannin; HT: hydrolysable tannin; M: myrobalan tannin; PGG: pentagalloylglucose; Q: quebracho; T: tannin of unknown origin; TA: tannic acid; W: wattle tannin.

pattern of B-ring hydroxylation of monomeric flavanols inhibitory than their catechin counterparts. If the toxicity has been shown to a!kct the level of growth inhibition of mechanisms of condensed tannins and flavanols are Stretptococcus mutans and Streptococcus sobrinus [34], related, we may deduce that prodelphinidins 2 will be Clostridium botulinum [24], Proteus vulgaris and more inhibitory than procyanidins 1, at least for these Stephylococus [44]: in all cases, gallocatechins were more bacterial specks.

3878 A. SCALBERT

The toxicity of tannins and lower M, phenols has also been discussed in relation to their oxidation state. Catechin was shown to be devoid of any toxicity against methanogenic bacteria, whereas if oxidized by oxygen in alkali, it strongly reduced methane production 1453. Similarly, bactericidal activity against Xanthomonas phaseoli of phenols such as catechol, protocatechuic acid or caffeic acid, is greatly enhanced upon oxidation by peroxidase [46]. These observations are possibly related to the effect of phenol oxidation with lactase or alkali on the inhibition of fungal enzymes such as pectinase, cellul- ase or peroxidase: oxidation increases inhibition of these last enzymes by catechol, protocatechuic acid or caffeic acid, but it decreases inhibition by a tannin (of unknown origin) [473. Red beet P-glucan synthase is also more strongly inhibited by various phenols once they have been oxidized by polyphenoloxidase; the effect of oxida- tion, although also positive for tannic acid, is less marked than for smaller phenols [48].

Two reasons may possibly explain this positive effect of phenol oxidation on enzyme inhibition. First, oxidized phenols may react with sulphydryl groups of enzymes and form covalent linkages with them [47, 481. Second, an increase of M, through oxidative polymerization [49], could also contribute to increase the efficiency of binding to proteins through non-covalent linkages (see below).

MECHANISMS OF TANNIN TOXICITY

Astringency: enzyme inhibition and substrate deprivation

If mechanisms of tannin toxicity are unique, they should be related to some of their characteristic physico- chemical properties, and in particular to their astrin- gency, as has often been claimed. Indeed, it has been found that addition of a ligand which will compete with microbial ligands such as enzymes can reduce inhibition by tannins. Addition of bovine serum albumin in the culture medium totally removed inhibition of Crinipellis

perniciosa germination by cocoa procyanidins [5]. The astringent character of tannins may induce com-

plexation with enzymes or substrates. Many microbial enzymes were found to be inhibited when raw culture filtrates or purified enzymes were mixed with tannins: cellulases [47,50,5 I], pectinases [52,53], xylanases [52], per- oxidase [47,52], lactase [52] or glycosyltransferase [32,33]. Natural polymers treated with tannins, were poorly degraded by microbial enzymes. Decomposition of glia- din, gelatin [54, 551 or polysaccharides such as rye hemicellulose, pectins or polygalacturonic acid [56], into CO,. and ammonia by a soil inoculum was strongly reduced by chestnut or wattle tannins when a sufficient (over I) tannin/substrate ratio is used. In comparison, decomposition of peptides (peptone) [54], aminoacids (phenylalanine [54] or lysine [SS]) and glucuronic acid [56] was much less affected. This is probably due to a lower affinity of these small molecules for tannins, as compared to polymers.

The same mechanism based on the astringent character of tannins may also explain the reduction of virulence of viruses such as tobacco mosaic virus [57, 581 or Herpes simplex virus [59]. The inactivation of tobacco mosaic virus by tannins is reversed by addition of gelatin [60] or of an alkaloid, nicotine sulphate [6l]. Astringency of tannins may also explain the reduction of larvicidal

activity of bacteria such as Bacillus thuringiensis, through interaction with their endotoxins [62].

Astringency of tannins is closely correlated with their chemical structure. An increase of protein binding effici- ency with M, of proanthocyanidins has commonly been observed [43,63-661. Dimeric ellagitannins are also more astringent than related monomers [42,43,67]. Affinity of gallotannins for bovine serum albumin increases with the number of galloyl residues [43, 681. Thus, if tannin toxicity is due to their astringent character, higher toxicity of higher molecular weight molecules should occur. Such a relationship has not yet been tested as few such toxicity studies have used purified, well-character- ized tannins. The relatively high toxicity of (-)-epicatechin gallate 10 and ( -)-epigallocatechin gall- ate 12 toward Clostridium botulinurn, as compared to their ungalloylated counterparts 9 and II, is possibly cx- plained by their high relative astringency [43].

However, this does not always occur. In some in- stances, the toxicity of tannins is no higher than that of catechins [ 131, although catechins have a very poor affinity for proteins [43,48,65]. It has also been observed that addition of bovine serum albumin to a glucosyl- transferase solution before addition of different gallotan- nins, failed to remove the inhibition of the enzyme by the tannins; the authors concluded that inhibition of the enzyme is not necessarily due to the non-specific binding of tannins to the enzyme [33].

Furthermore, it has not yet been established so far, which of two mechanisms, inhibition of extracellular enzymes or deprivation of substrates, is effectively in- volved in tannin toxicity. Tannins may also directly affect the metabolism of microorganisms, as suggested by modi- fication of the morphology of the germ tube of Crinipellis perniciosa at low tannin concentration (0.063 g 1.. I) [S]. A drastic change in the morphology and growth pattern of bacteria was observed when they were grown in the presence of subinhibitory concentrations (0.6. I g I _ ‘) of tannic acid or carob pod extract [37]: Pseudomonas

jluorescens, Escherichia coli or Cellcibrio jiilws formed chains or filaments whereas most cells were single when grown in the absence of tannins; the morphology of other bacterial species, although also subject to tannin inhibi- tion, was not affected.

Action on membranes

Different mechanisms may induce such effects on the metabolism of microorganisms. First, they may be medi- ated through action on their membranes. Indeed. tannic acid at low concentration (less than 1 mgl ‘) inhibits oxidative phosphorylation by mitochondria of blowfly flight muscle [69]. Inhibition of the electron transport system was also observed on rat liver mitochondria with a tannic acid concentration of 50 mg I _ ‘: integrity of the membranes was affected at twice this concentration [70]. Similar mechanisms were shown to be involved in the antibacterial properties of tannic acid on Photohncterium phosphoreum [29]. The action of tannins on bacteria could thus be similar to that of related synthetic phenols such as diphenyl (e.g. o-phenyl-phenol) and diphenylalk- ane (e.g. hexachlorophene) compounds which have found wide application as disinfectants [71]. It should be noted that to reach the membrane, the tannins must cross the microbial cell wall. This cell wall made of different polysaccharides and proteins probably fix part of the

Ahmicrobial properties of tannins 3879

tannins. This fixation could contribute to an increase in the minimum inhibitory concentration as compared to the values given above for inhibition of mitochondrial respiration. Such a dependence on adsorption was ob- served with synthetic phenolic antibacterial compounds as their MIC increases with inoculum size [71].

Metal ions deprivation

Another toxicity mechanism might involve com- plexation of metal ions by tannins. Biological systems including microorganisms are highly dependent on the metal ion status of the environment [72]. Antimicrobial activity through iron depletion is particularly well docu- mented [73]. For example, infection of humans by Escherichia co/i is inhibited by the iron chelating lacto- ferrin present in human milk, but restored by an excess of iron [743. Lactoferrin is also released in septic areas by degranulation of blood circulating leukocytes [75]. Con- albumin, another iron binding protein, accounts for 12% of the egg-white solids and prevents microbial invaders reaching the chicken yolk [73]. Depletion of metal ions through precipitation was postulated as a mode of fungi- tidal action of hydrogen sulphide which would form insoluble metal sulphides [76]. Tannins might exert their antimicrobial action through a similar mechanism. In- deed, most of the tannins have more than two o-diphenol groups in their molecule, which can form chelates with many metal ions such as ferric or cupric ions [77]. The multi-catecholate nature of tannins, which allows reticul- ation, is probably at the origin of the formation of metal- tannin precipitate (Fig. 2) [78, 791. Such an original property of tannin which makes them different from smaller phenols has been known for centuries and has been used in the making of inks of good washing fastness.

It is thus possible that tannins reduce availability of essential metal ions for microorganisms, just as tannins reduce iron absorption in rats given tannin-rich be- verages like tea or cocoa [80]. Similarly, ionizable iron in tannin-rich varieties of ragi (Eleusine coracana) is deple- ted, as compared to other varieties containing no tannins [81]. Metal depletion by tannins may affect the activity of metalloenzymes. Indeed, tannins were shown to be parti- cularly good inhibitors of fungal metalloenzymes such as peroxidase and lactase as compared to other enzymes such as cellulase, pectinase, xylanase or protease [SZ]. Other chelating molecules such as tropolone [SZ] or DIECA [83] are also known to inhibit copper-containing polyphenol oxidases. The incidence of such an inhibition on microbial metabolism is, however, not ascertained, as

Fig. 2. Insoluble tannin-ferric ion complexes

h

involvement of these enzymes in plant infection or plant material degradation is poorly understood.

MICROBIAI. DEFENCES AGAINST TANNIUS

Despite the antimicrobial properties of tannins, many microorganisms can grow and develop on tannin-rich materials. Many woods such as those of pedunculate oak or sweet chestnut contain over 10% of their dry weight as tannins and are among the most durable organic mater- ials that living organisms synthesize [7]. None of them is, however, non-biodegradable. Some moulds develop easily on the surface of tannin-rich woods such as que- bracho [84] or European oak (personal observation). Moulds such as Aspergillus niger or Pencillium ylaucum grow on the surface of the liquid of tannery pits [84]. Several detoxification mechanisms may contribute to the explanation of the growth of microorganisms on tannin rich media.

Secretion of tannin-binding polymers

Microorganisms may secrete outside the cell, polymers with a high affinity for tannin with which they will combine. Tannins will thus become unavailable for com- bination with other molecules such as microbial enzymes, which are essential to the growth of the microorganism. Condensed tannins induce the secretion of a water- soluble mucilage by spores of Colletorrichum graminicola [ 143. This mucilage is made of glycoproteins which have a high affinity for condensed tannins. Similarly, bacteria from rumen fluid, when grown on a high-tannin variety of birdsfoot trefoil (Lotus cornicularus), produce a glyco- calyx which is absent when the bacteria are grown on a low-tannin variety of the same legume species [41]. This mechanism of tannin detoxification is probably wide- spread in biological systems. The same strategy is used by plants and animals to reduce astringency of tannins. The reduction of astringency in ripening fruits does not seem, at least in some fruits, to depend on a reduction of the amounts of tannin, but on the release of tannin-binding molecules [SS, 863. Some animals like rats are able to develop normally on tannin-rich diets. Such diets were shown to induce in these animals secretion of hydroxy- proline-rich salivary proteins with a high affinity for tannins [87].

Tannin-resistant enzymes

Although many enzymes are inhibited by tannins as previously stated, a few enzymes such as tannases or an invertase [SS] and an a-amylase [89] isolated from seeds of Leoti sorghum and Caluatia gigantea, respectively, are known to maintain full activity in the presence of high concentrations of tannins. Apart from the obvious case of tannases, the extent to which the presence or absence of such enzymes contribute to the success or failure of microorganisms to develop on tannins is not known.

Tannin oxidation

Oxidation of tannins by polyphenoloxidases has also been inferred in studies of their detoxification. Tannins induce the secretion of lactase by fungi such as Trumetes hirsuta [19]. Fames annosus laccases [I73 and Xantho- monas mulr;acearum peroxidase [90] were similarly in-

3880 A. SCALBERT

Table 2. Micro-organisms capable of growing on tannins as a unique source of carbon

Mtcroorganisms Tannin substrate Ref.

Filamentous fungi

Asprrgillus niqer

Asprrgillus niyer

Cahatia yiganfra

Penicillium sp.

Penicillium adametzi

Trwhoderma riride

Various soil fungi

Various fungi

Yeasts

Candida @hermondii

Candida tropicalis

Torulopsir candtda

Saccharomyces rouvii

Various yeast species from tannery

liquors and xylophagc insects

Bacterta

Achromobacter sp.

Azotobacter vinelandii

Escherichia coli

Pseudomonas jluorescens

Corynebacterlum sp. Klebsiella pneumoniae

Various bacteria from forest soil

Various bacteria from forest soil

Extract from M?:rtus communis seeds W

Ch. W

T

Purified oligomeric CT

cl-

Ch, GT. W

TA

W Extract from strawberry rhizome

Ch, Q. W

Ch, GT

Ch, W

Tara T

TA

Ch

WI Cl21

[971

[9gl

L991 [201

JtmJ [loll

Cl021 L to31

~233

Cl~l

1351

[l@tl Cl051 CtOoJ

Abbreviations: see Table I

duced by smaller phenols such as benzoic acids. cinnamic acids or catechin. However, the physiological role of these oxidascs is far from evident. No clear relationship could be established between extracellular laccases and the toxicity of phenols [17]. Laccases may oxidise tannins and thus affect their toxicity [52]. They may also induce the co-polymerization of soluble proteins with tannins or other phenols and thus eventually inactivate fungal en- zymes to the detriment of the invader [913.

Sidcrophores

Another possible way for microorganisms to deter tannin defences of plants may be related to their ability to sequester metal ions. Egg white [92] and cheeses [933 containing conalbumin and lactoferrin respectively, which both contribute to reduce the availability of iron, were shown to be enriched in microbial siderophores when heavily contaminated by fungi and bacteria. Wood- decaying basidiomycetes also produce siderophores which may allow the fungi to accumulate adequate iron [94]. The presence of these siderophores has not yet been examined in relation to the concentration of tannins in woods, but it is possible that a high amount of tannins will induce a strong requirement for siderophores to assimilate the 10 ppm (based on dry wood) [953 of iron generally present in woods.

Tannin biodegradation

Some microorganisms grown in pure culture were shown to develop on media containing tannins as sole

source of carbon (Table 2). Both condensed and hydro- lysable tannins have been used as substrates. Several kinds of tannins have been compared. Aspergillus, Peni-

cillium, Fomes, Polyporus, Poria and Trametes species were shown to grow better on a gallotanin than on chestnut tannin (ellagitannin) or wattle tannin (conden- sed tannin) whereas a bacteria, Achromobacter sp., was unable to grow on catechin or wattle tannin but could grow on gallotannin and chestnut tannin [IOO].

Most of these microorganisms which grow on tannin are also able to grow on simpler phenols such as gallic acid or (+)-catechin [20,97]. The reverse is not obvious. Penicillium sp. which degrade catechin show limited or no growth on wattle tannin [ 1001. The growth of Penicillium adametzi on purified flavanoids of varying M, has been examined: mycelium growth was maximal on (+)- catechin and decreased on the following substrates: dimer B-3, trimer C-2 and a condensed tannin fraction from Pinus radiata bark [99]. These observations tend to confirm that unique mechanisms contribute to tannin toxicity.

Limited data exist on the tannin biodegradation path- ways used by microorganisms. Degradation of hydro- lysable tannins, and particularly gallotannins, is best understood. Some fungi have been known for a long time to produce an enzyme, tannase, which allow them to grow on hydrolysable tannins. Over 20 filamentous fungi were tested for growth on tannic acid. Although most of them were able to grow under a low concentration of tannic acid (2.5 g I- I), only a few Penicillium and all the Aspergillus species tested showed good growth under concentrations of tannic acid as high as IOOgI-r [loll.

Antimicrobial properties of tannins 3881

These last fungi produce a tannase which is induced by gallate methyl ester [ 1073, tannic acid [ 1023 but not by other simple phenols such as gallic acid, salicylic acid or salicylate methyl ester [102]. The enzyme activity is partly soluble and partly insoluble, bound to the mycel- ium [ 102,108,109]. Its pH optimum for activity is 5.0-6.0 [ 110, 11 l] and it is unstable above pH 6 [ 1111.

Tannase catalyses the hydrolysis of ester bonds be- tween a phenolic acid and an alcohol. The nature of the alcohol in the ester bound has little intluence on the activity. It can be methanol [ 1073, glucose [ 1121, gluconic acid [113], quinic acid [114, 1151 and other related alicyclic compounds [116], flavanols [117, 1181, gallic acid (the ester bond is a depside as in 3) Cl073 or a terpene [119]. The requirement on the carboxylic acid is stronger. It must be benzylic and must belong to a phenolic acid carrying two hydroxyl groups in ortho position as in protocatechuic or gallic acid 7 [l lo]. The extensive use of a commercial tannase [ 1203 or a tannase- rich raw extract from Aspergilhs niger Cl213 in structural studies of hydrolysable tannins by Nishioka’s and Okuda’s groups respectively, has led to a better under- standing of their precise steric requirements on the car- boxylic acid. Tannase is active on galloyl residues of galloyl esters, as well as on hexahydroxydiphenoyl [ 122, 1231, dehydrodigalloyl [123] and valoneoyl [124] re- sidues of ellagitannins. Galloyl residues are usually more easily hydrolysed than the other groups [125,126]. Their hydrolysis will depend on the position of the substituents on the glucose. 1-0-Galloyl residues will usually be hydrolyzed first, followed by 2-0-galloyl residues and lastly 3-0-galloyl residues [127, 1281. 4,6-O-Hexa- hydroxydiphenoyl residues are more easily hydrolysed than the 2,3-O-hexahydroxydiphenoyl residues [ 122, 1231. The duration of hydrolysis can be varied from several minutes to several hours, and analysis of the different products can be used as an aid for structure determination of hydrolysable tannins [ 125, 1291.

Several attempts have been made to produce or isolate different tannases from the culture media of Aspergillus, and in particular to separate the esterase activity catalys- ing the hydrolysis of galloyl esters on glucose, from the depsidase activity catalysing the hydrolysis of ester lin- kages between two galloyl residues [107, 1303. These attempts confirmed that this fungus produces several tannase isoenzymes, but failed to provide fractions with exclusive esterase or depsidase activity. The ratio of the two activities did vary in the different fractions but the relative specificity of each enzyme was low. Tannase activities were also reported in bacteria [106, 1 lo] and yeast [23, 1313 but have not been characterized in detail.

Gallic acid produced by microorganisms may accumu- late or be further degraded. Moulds growing on the surface of gallotannin containing tannery pits do not degrade gallic acid if the solutions are mixed every day. On the other hand, if the fungi are allowed to grow freely on the surface, most of the gallic acid will be degraded; up to 25 g of Aspergillus mycelium and large amounts of CO, will be produced from 80 g of tannin [84]. This is possibly explained by the difference in oxygen supply which is required for gallic acid degradation [loll. Pathways of gallic acid degradation have been deter- mined [132, 1333. On the other hand, the pathways leading to the degradation of hexahydroxydiphenoyl groups 4 (or ellagic acid, 13) and related biphenyl and biarylether structures have not been examined.

Biodegradation of proanthocyanidin is much less well understood. Depolymerization by yeasts has been sugges- ted [103]. Their degradation may take in part the path- ways identified for the degradation of ( + )-catechin [ 1341 and other monomeric flavanols [132, 135, 1361.

The implication of tannin biodegradation in the suc- cess or failure of microorganisms to develop on tannin- rich materials is not known. Degradation of tannins may be required to make substrates such as proteins or polysaccharides available. It may also contribute to remove metal ions such as iron from their complexes, just as soil microorganisms degrade plant siderophores [137]. However, biodegradation of tannins may not be a pre- requisite to develop on tannin-rich materials for micro- organisms equiped with other means of defence such as secretion of polymers which will combine with the tan- nins.

CONCLUSIONS

It appears that there are probably several mechanisms involved in tannin toxicity. The obvious interactions of tannins with enzymes or substrates may have led to a failure to examine other possible mechanisms involved in their toxicity, such as direct action on membranes or deprivation of iron. Their mode of action probably depends on the individual microorganism. This could explain the large differences in MIC values between fungi, bacteria or yeasts. Studies using purified tannin molecules instead of crude extracts, are still few in number and will undoubtedly lead to a better understanding of their antimicrobial properties.

Nevertheless, even if some of the particular physico- chemical properties of tannins seem to be effective in plant defence, in other cases, tannins are no more toxic than the smaller non-astringent related phenols. The reason why plants have selected tannins for their defence should thus be looked for elsewhere, and possibly in the plant itself. Tannins can form highly concentrated solu- tions. This property has been interpreted as mutual solubilization [ 1381, which is probably explained by the characteristic macromolecular properties of tannins. Such a property may have benefited plants which could accumulate large quantities of polyphenols in their vacuoles. These large concentrations of tannins would counterbalance their low toxicity for microorganisms.

Several (if not many) microorganisms have however evolved to withstand these high concentrations of tan- nins. Some of them have succeeded particularly well. Fungi such as Aspergillus or Pencillium spp. which pro- duce tannases are good examples. Other fungi, bacteria or yeasts are possibly as well equipped but the data are still very scarce. The strategies which microorganisms have developed in the course of evolution, are probably various. We may expect modes of defence to be specific for a microorganisms species or family, just as rats and mice differ from hamsters in their ability to secrete hydroxyproline-rich salivary proteins and thus to with- stand tannin-rich diets [ 1393.

Acknowledgements-The author thanks Dr B. Kurek, Dr B. Monties and Dr D. A. Wood for helpful comments on the manuscript.

3882 A. SCALBERT

REFERENCES

1. Spencer, C. M., Cai, Y., Martin, R., Gaffney, S. H.,

Goulding, P. N., Magnolato, D., Lillcy, T. H. and Haslam,

E. (1988) Phytochemistry 27, 2397. 2. Haslam, E. (1989) PIant Polyphenols. Cambridge University

Press, Cambridge. 3. Feeny, P. (1970) Ecology 51, 565.

4. Jersch, S., Scherer, C., Huth, G. and Schliisser, E. (1989)

Z. Pjlanzenkr. Pjlanzensch. 96, 365.

5. Brownlee, H. E., MC Euen, A. R., Hedger, J. and Scott, 1. M.

(1990) Physiol. Molec. Plant Pathol. 36, 39.

6. Hart, J. H. and Hillis, W. E. (1972) Phytopathology 62,620.

7. Scalkrt, A. (1992) Proceedings of the 2nd North American

Tannin Conference. Plenum Press, New York. 8. Laks, P. E. (1987) Phyrochemistry 26, 1617.

9. Laks, P. E., McKaig, P. A. and Hemingway, R. W. (1988)

Hoizforschuny 42, 299.

10. Smith, A. L., Campbell, C. L.. Diwakar, M. P., Hanover,

J. W. and Miller, R. 0. (1989) Holzforschung 43, 421. Il. Kawamura, J. and Takeo, T. (1989) J. Jap. Sot. Food Sci.

Trchnol. 36. 463.

12. Sivaswamy, S. N. and Mahadevan, A. (1985) J. Environ.

Biol. 6, 27 1. 13. Sivaswamy, S. N. and Mahadevan, A. (1986) 1. Indian Bot.

Sot. 65, 95. 14. Nicholson, R. L., Butler. L. G. and Asquith, T. N. (1986)

Phytopathology 76, I3 15. 15. Rudman, P. (1962) Holzforschung 16, 56.

16. Wang, S., Hart, J. and Behr, E. A. (1980) For. Prod. J.30.55.

17. Haars. A., Chet, 1. and Hiittermann, A. (1981) Eur. J. forest

Pa&o/. 11, 67. 18. Wehmer. C. (1912) Mycol. Cenrr. 1, 138.

19. Arora. D. S. and Sandhu, D. K. (1984) Folio Microbial.

(Pruyue) 29, 3 IO.

20. Arrieta-Escobar. A. and Behn, J. M. (1982) Eiotechnol.

Bioenyng 24. 983.

21. Harris, H. B. and Burns, R. E. (1973) Agron. J. 65, 957.

22. Mullins, J. T. and Nesmith, C. (1988) Biomass 16. 77.

23. Jacob, F. H. and Pignal, M. C. (1972) Mycopathol. Mycol.

Appl. 48, 121.

24. Hara, Y. and Watanabe. M. (1989) J. Jap. Sot. Food Sci.

Technol. 36, 95 I. 25. Farrer, T. W., Biek, L. and Wormwell, F. (1953) 3, 80.

26. Waage, S. K., Hedin, P. A. and Grimley, E. (1984) Phyto-

chemlsfry 23, 2785.

27. Rice, E. L. and Pancholy, S. K. (1973) Am. J. Botany 60.691.

28. Nishizawa, K.. Nakata, I., Kishida, A., Ayer, W. A. and

Browne, L. M. (1990) Phytochemisrry 29. 2491. 29. Konishi, K., Adachi, H., Kita, K. and Horikoshi, 1. (1987)

Chem. Pharm. Bull. 35, 1169. 30. Muthukumar, G. and Mahadevan, A. (1981) Indian J. Exp.

Eiol. 19, 1083.

31. Sening, E. (1963) Zentr. Bakteriol. Parasitenk--II 117,

13. 32. Hada, S., Kakiuchi, N., Hattori, M. and Namba, T. (1989)

Phyrother. Res. 3, 140. 33. Kakiuchi, N., Hattori, M., Nishizawa, M., Yamgishi, T.,

Okuda, T. and Namba. T. (1986) Chem. Pharm. Bull. 34,

720.

34. Sakanaka, S., Kim, M.. Taniguchi, M. and Yamamoto, T.

(1989) Agric. Biol. Chem. 53, 2307.

35. Basaraba, J. (1966) Can. J. Microbial. 12, 787.

36. Baldwin, I. T.. Olson, R. K. and Reiners, W. A. (1983) Soil

Biol. Biochem. IS, 419.

37. Henis, Y., Tagari, H. and Volcani, R. (1964) Appl. Microbial.

12, 204.

38. Hara, Y. and Ishigami. T. (1989) J. Jap. Sot. Food. Sci.

Technol. 36. 996.

39. Toda, M., Okube, S., Hiyoshi, R. and Shimamura, T. (1989)

Lett. Appl. Microbial. 8, 123.

40. Field, J. A. and Lettinga, G. (1987) Wafer Res. 21, 367.

41. Chiquette, J., Cheng, K. J., Costerton, J. W. and Milligan, L. P. (1988) Can. J. Anim. Sci. 68, 751.

42. McManus. J. P., Davis. K. G., Beart, J. E., Gaffney, S. H.,

Lilley, T. H. and Haslam, E. (1985) J. Chem. Sot. Perkin

Trans II 9, 1429.

43. Okuda, T., Mori, K. and Hatano. T. (1985) C/rem. Pharm.

Bull. 33, 1424.

44. Mori, A., Nishino, C., Enoki. N. and Tawata. S. (1987) Phytochemistry 26. 2231.

45. Field. J. A., Kortekaas, S. and Lettinga. G. (1989) Biol.

Wastes 29. 241.

46. Urs, N. V. R. R. and Dunleavy. J. M. (1975) Phyro-

pathology 65, 686.

47. Lyr. H. (1965) Phytopathol. Z. 52, 229.

48. Mason, T. L. and Wasserman. B. P. (1987) Phytochemistry

26, 2197.

49. Hathway, D. E. and Seakms, J. W. T. (1957) Biochem. J. 67,

239.

50. Benoit, R. E. and Starkey, R. (1968) Soil Sci. 105, 203.

51. Mole, S. and Waterman, P. G. (1987) Oecologia 72. 148.

52. Lyr, H. (1961) Enzymoloyia 23, 231.

53. Williams A. H. (1963) in Enzyme Chemistry (4 Phenolic

Compounds (Pridham J. B.. ed.), p. 87. Pergamon Press, New York.

54. Basaraba, J. and Starkey, R. L. (1966) Soil SCI. 101, 17.

55. Benoit, R. E., Starkey, R. L. and Basaraba. J. (1968) Soil. Sci.

105, 153.

56. Benoit, R. E. and Starkey, R. L. (1968) Soil Sci. 105. 291.

57. Bawden. F. C. and Kleckowski, A. (1945) J. Porno/. 21, 2.

58. Van Schreven. D. A. (1941) h’aruurwer. Tijdschr. Ned.-lnd.

4, 113.

59. Takechi, M., Tanaka, Y., Takehara, M., Nonaka, G.-I. and Nishioka, I. (1985) Phytochemistry 24, 2245.

60. Thornberry, H. H. (1935) Phytopatholoyy 25, 931.

61. Thung, T. H. and Van Der Want. J. P. H. (1951) Tijdschr.

Plantenziekten 57. 173.

62. Lord, J. C. and Undeen. A. H. (1990) Enriron. Entomol. 19,

1547.

63. Porter, L. J. and Woodruffe. J. (1984) Phytochemisfry 23,

1255.

64. Kumar, R. and Horigome, T. (1986) J. Agric. Food Chem.

34,487.

65. Ezaki-Furuichi, E.. Nonaka, G. I.. Nishioka. 1. and

Hayashi, K. (1987) Ayric. Biol. Chem. 51. I IS.

66. Artz, W. E., Bishop. P. D., Dunker, A. K., Schanus. E. G.

and Swanson, B. G. (1987) J. Agric. Food Chem. 35. 417.

67. Herve du Penhoat. C. L. M.. Michon, V. M. F.. Peng. S.,

Viriot, C.. Scalbert, A. and Gage, D. (I 99 I) J. Chem. Sot.,

Perkin Trans I 1653.

68. Verzele. M., Delahaye, P. and Van Damme. F. (1986)

J. Chromatogr. 362, 363.

69. Duncan, C. J., Bowler, K. and Davidson, T. F. (1970)

Bfochem. Pharmacol. 19. 2453.

70. Adachi. H., Konishi. K., Toriizuka. K. and Horikoshi. I.

(1987) Chem. Pharm. Bull. 35. I 176.

71. Hamilton, W. A. (1971) in Inhibition and Destruction o/the

Microbial Cell (Hugo. W. B.. ed.), p. 77. Academic Press,

London.

72. Brown, J. C. (1963) Annu. Rec. Plant Physiol. 14, 93.

73. Weinberg, E. D. (1984) Phpsiol. Rec. 64, 65.

74. Bullen. J. J., Rogers H. J. and Leigh, L. (1972) Brrr. Med. J.

I, 69.

Antimiaobial properties of tannins 3883

75. Van Snick, J. L., Masson, P. L. and Heremans, J. F. (1974)

J. Exp. Med. 140, 1068.

76. Horsfall, J. G. (1956) Principles of Fungicidal Action. Chronica Botanica, Waltham.

77. Harris, C. M. and Livingstone, S. E. (1964) in Chelating Agents and Metals Chelates. (Dwyer, F. P. and Mellor,

D. P., eds), p. 95. Academic Press, New York.

78. Okuda, T., Mori, K., Shiota, M. and Ida, K. (1982) Yaku-

gaku Zasshi 102, 735.

79. Kennedy, J. A. and Powell, K. J. (1985) Aust. J. Chem. 3&

879.

80. Brown, R. C., Klein, A., Simmons, W. K. and Hurrell, R. F.

(1990) Nutr. Rex 10, 343.

81. Udayasekhararao, P. and Deosthale, Y. G. (1988) Plant Foods Hum. Nurr. 38, 35.

82. Kahn, V. and Andrawis. A. (1985) Phytochemistry 24,905,

83. Pierpoint. W. S. and Harrison. B. D. (1963) J. Gen. Micro-

biol. 32, 429.

84. Meunier, L. and Vaney. C. (1903) La tannerie. Gauthier-

Villars, Paris.

85. Joslyn, M. A. and Goldstein, J. L. (1964) Adu. Food Res. 13.

179.

86. Ozawa, T., Lilley, T. H. and Haslam, E. (1987) Phyto-

chemistry 26, 2937.

87. Mehansho, H., Butler, L. G. and Carlson, D. M. (1987)

Annu. Rev. Nulr. 7, 423. 88. Strumeyer, D. H. and Malin, M. J. (1970) Biochem. J. 118,

899. 89. Komninos, J., Kekos. D., Macris, B. J. and Galiotou-

Panayotou, M. (1988) Biotechnol. Bioeng. 32, 939.

90. Venere, R. J. (1980) Plant Sci. Lerters 20.47.

91. Leatham, G. F., King, V. and Stahman, M. A. (1980)

Phytopathology 70, 1134. 92. Wawszkicwicz, E. J. (1974) Microbiology 299.

93. Ong, S. and Neilands, J. B. (1979) J. Agric. Food. Chem. 27, 990.

94. Fekete. F. A., Chandhoke, V. and Jellison, J. (1989) Appl. Environ. Microbial. 55, 2720.

95. Fengel, D. and Wegener, G. (1984) Wood, Chemistry, Ulrrastructure, Reactions. Walter de Gruyter, Berlin.

96. Guinea-Lopez, M. C. and Ortega, M. (1985) An. R. Acad. Farm. 51, 347.

97. Galiotou-Panayotou, M. and Macris, B. (1986) J. Appl. Microbial. Biotechnol. 23, 502.

98. Cowley, 0. T. and Whittingham, W. F. (1961) Mycologia

53, 539.

99. Grant, W. D. (1976) Science 193, 1137.

100. Lewis, J. A. and Starkey, R. L. (1969) Soil Sci. lM, 235. 101. Knudson, L. (1913) J. Biol. Chem. 14, 159.

102. &iik, G. and Deschamps, A. M. (1983) Mycopathologia 83, 107.

103. Vennat, B., Pourrat, A. and Pourrat, H. (1986) J. Ferment.

Tech&. 64, 227

104. Deschamps, A. M. and Leheault, J. M. (1984) Biotechnol.

Letters 6, 237. 105. Deschamps, A. M., Mahoudeau, G., Conti, M. and

Leheault. J. M. (1980) J. Ferment. Technol. !%, 93. 106. Deschamps, A. M., Otuk, G. and Leheault, J. M. (1983)

J. Ferment. Technol. 61, 55. 107. Haslam, E. and Stangroom, J. E. (1966) Biochem. J. 99.28.

108. Freudenherg, K., Bliimmel, F. and Frank, Th. (1927)

Hoppe-Seyler’s 2. Physiol Chem. 164, 262.

109. Jean, D., Pourrat, H. and Pourrat, A. (1978) Bull. Liaison

Groupe Polyphbnols 134.

110. Dyckerhoff. H. and Armbruster, R. (1933) Hoppe-Seyler’s Z. Physiol. Chem. 219. 38.

111. Thomas, R. L. and Murtagh, K. (1985) J. Food Sci. 50.1126.

112. Haslam, E., Haworth, R. D., Mills, S. D., Rogers, H. J.,

Armitage, R. and Searle, T. (1961) .I. Chem. Sot. 1836.

113. Tanaka, T., Nonaka, G. and Nishioka, 1. (1986) Chem.

Pharm. Bull. 34, 656.

114. Nishimura, H., Nonaka, G.-I. and Nishioka, I. (1984)

Phytochemistry 23, 2621. 115. Pourrat, H., Regerat, F., Morvan, P. and Pourrat, A. (1987)

Biotechnol. Letters 9, 731.

116. Ishimaru, K., Nonaka, G.-I. and Nishioka, 1. (1987) Phyto-

chemistry 26, 1501.

117. Nonaka, G., Nishioka, I., Nagasawa, T. and Oura, H.

(1981) Chem. Pharm. Bull. 29, 2862.

118. Hashimoto. F., Nonaka, G. and Nishioka, I. (1989) Chem.

Pharm. Bull. 37. 3255.

119. Hatano, T., Yasuhara, T., Abe, R. and Okuda, T. (1990)

Phytochemistry 29, 2975.

120. Nonaka, G., Nishimura, H. and Nishioka, I. (1985) J. Chem.

Sot., Perkin Tram I 163. 121. Yoshida,T., Tanaka, K.,Chen, X.-M. and Okuda, T. (1989)

Chem. Pharm. Bull. 37,920.

122. Schmidt, 0. T.. Wiirtele, L. and Harrtus, A. (1965) Lieb.

Ann. Chem. 690,150.

123. Yoshida, T., Okuda, T., Memon, M. U. and Shingu, T.

(1985) J. Chem. Sot., Perkin Trans I 315.

124. Hatano, T., Kira, R., Yasuhara, T. and Okuda, T. (1988) Ckm. Pharm. Bull. 36, 3920.

125. Nonaka, G., Tanaka, T. and Nishioka, 1. (1982) J. Chem.

Sot., Perkin Trans I 1067.

126. Nonaka, G., Nakayama, S. and Nishioka, 1. (1989) Chem. Pharm. Bull. 37. 2030.

127. Xin-Min. C., Yoshida, T., Hatano, T., Fukushima, M. and

Okuda, T. (1987) Phytochemistry 26, 5 15.

128. Hatano, T., Yashuara, T. and Okuda, T. (1989) Chem. Pharm. Bull. 37, 2665.

129. Nonaka, G., Ishimaru, K., Azuma, R., Ishimatsu, M. and

Nishioka, 1. (1989) Chem. Pharm. Bull. 37, 2071.

130. Beverini. M. and Metche, M. (1990) Sci. Alim. 10, 807. 131. Aoki, K., Shinke, R. and Nishira, H. (1976) Agric. Biol.

Chem. 40.79.

132. Barz, W., Weltring, K.-M. (1985) in Biosynthesis and Bio-

degradation of Wood Components. (Higuchi, T., ed.), p. 607.

Academic Press, Orlando.

133. Watanabe, A. (1965) Agric. Biol. Chem. 29, 20. 134. Chandra, T., Madhavakrishna, W. and Nayudamma, Y.

(1969) Can. J. Microbial. 15, 303.

135. Barz, W. and H&l. W. (1975) in 7%e Flauanoids, (Harhorne, J. B., Mabry, T. J. and Mabry, H., eds), p. 916.

Chapman & Hall, London. 136. Winter, J., Moore, L. H., Dowell, V. R. Jr. and Bokken-

heuser, V. D. (1989) Appl. Environ. Microbial. 55, 1203.

137. Romheld, V. and Marschner, H. (1986) Plant Physiol. So, 175.

138. White, T., Kirby, K. S. and Knowles, E. (1952) J. SOC.

Leather Trade Chem. 36, 148.

139. Butler, L. G., Rogler, J. C., Mehansho, H. and Carlson, D. M. (1986) in Plant FIauonoids in Biology and Medicine:

Biochemical, Pharmacological and Structure-Activity Rela- tionships, p. 141. Alan R. Liss.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具