Camp.Biochem.Physiol.Vol. IOIA,
No.
0300-9629/92$5.00+ 0.00 0 1992Pergamon Press plc
2, pp. 381-385,1992
Printed in Great Britain
THE USE OF PURIFIED CONDENSED TANNIN AS A REFERENCE IN DETERMINING ITS INFLUENCE ON RUMEN FERMENTATION W.
VAN
HOVEN* and D. FUR~TENBURG
Centre for Wildlife Research, University of Pretoria, Pretoria 0002, South Africa. Telephone: (0 12) 420-2627 [Received 29 April 1991) Ah&net-I. Tannins were purified from the leaves of trees forming part of giraffe’s diet in the Kruger National Park. 2. In general, hydrolysable tannin formed less than 10% of the total purified crystal complexes, condensed tannin and possibly some non tannin phenols forming the balance. 3. Bach tree species condensed tannin gave a different callibration curve and these were used in the assay. 4. Volatile fatty acid production during in vitro fermentation was greatly reduced when the substrate contained more than 6% condensed tannin.
INTRODIJCI’ION
The role of tannins as a chemical deterent to herbivory has been well documented (Arnold and Hill, 1972; Freeland and Janzen, 1974; McLeod, 1974; Feeny, 1976; Rhoades and Cates, 1976; Swain, 1977; Haukioja, 1980; Touze and Esquerre-Tugaye, 1980; Rhoades, 1983,1985; Zucker, 1983; Beart et al., 1985; Cooper and Gwen-Smith, 1985; Cooper et al., 1988; Feath and Bultman, 1986; Robbins et al., 1987a,b; Waring and Price, 1988; Kearsley and Whitham, 1989). The term tannin is a general one for a large variety of combinations of basic phenolic (C,) molecules. The flavonoids are C&& molecules and their derivatives. The two major types of tannin (condensed and hydrolysable) are chemically quite different. Condensed tannins are divided into (A) proanthocyanidins which are di-, tri- or polymers of the anthocyanidin and or catechin flavan-3-01s and (B) leucoanthocyanidins which are dimers of the flavan3,4-diol flavonoids (Geissman and Hinreiner, 1952; Harborne, 1967; Sarkar et al., 1976; Rhoades and Wooltorton, 1978; Hagerman and Butler, 1989). Catechin is probably the best known due to its wide use as a standard in the vanillin assay of tannin. Hydrolysable tannins are gallic or hexahydroxydiphenic acid esters of glucose or other polyols (Haslam, 1977). The binding of proteins to tannins varies according to the types of proteins and tannins. The assays presently available measuring protein for binding/precipitation are not satisfactory in all respects (Asquith and Butler, 1985; Hagerman and Butler, 1989; Mole et al., 1989). Asquith and Butler (1985) showed that five different proteins varied by as much as three orders of magnitude in their relative affinities for sorghum grain tannin. Thus, owing to the large variety of anthocyanidins and condensed *To whom correspondence
should be addressed.
tannins that occur in plants, the use of catechin or any other single purified tannin as a standard in calibration could lead to inaccuarate estimates as shown by Wisdom et al. (1987). In this study a colometric assay based on the original description of Bums (1971) was used. Tannin
extracts were made of each plant to be analysed and the purified tannin thus obtained was used to set up a calibration curve for each particular plant species. When using catechin for a standard calibration curve, tannin values in grain sorghum were over estimated by two-fold at AA,, for absorbance values lower than 0.2 and by 3-fold for values higher than 1.4 (Hagerman and Butler, 1978; Price et al., 1978). In studying the influence which tannins have on the feeding, digestion and ecology of browsing ruminants, a reliable practical analytical procedure had to be established. Purifying a particular plant species tannin and using it to set up a calibration curve is advantageous in that this curve can be used for all further analyses of that particular plant species, provided the same asssay is used. Results obtained by means of this analytical procedure were related to the volatile fatty acid energy that becomes available as metabolizable energy during fermentation. MATERIALS
AND
METHODS
Based on the method of Strumeyer and Malin (1975) a 2.0 x 28.0 cm column was packed with 12 g Sephadex LH20 gel saturated with 99% methanol. Leaves of trees that occur in the diet of giraffe were collected, dried and ground in a laboratory mill. A log sample was washed four times in 60ml of peroxide-free ether by shaking the flasks at regular intervals over 3 hr and finally centrifuging, thus effectively removing the fats, fatty acids, essential oils and most chlorophyll. The sediment containing the flavonoids was resuspended in 200 ml 99% methanol-and left air sealed in the dark at room temperature for 48 hr. This 200 ml flavonol extract was filtered throuah the gel column and rinsed with an additional 80ml of&e
381
382
W.
VAN HOVEN and
methanol, thus removing most of the non-tannin flavonoids from the gel (Renas ef al.. 1969; Strumever and Malin. 1975). The-tannins precipitated to the gel were removed by rinsing the column with 70 ml eluent of 50% aqueous acetone. The filtrate was evaporated to the remain of dark rust-brown tannin crystals. Crystalixed tannin was immediately redissolved in 99% methanol in a ratio of 75.0 mg: 10.0 ml and stored air sealed at -4°C in the dark until the calibration dilutions were made based on Bums (1971). The dilution series started with a concentration of 1.125 mg tannin per ml methanol. Vanillin (2.500 g) in 250.0 ml 99% methanol was added to 60.0 ml 32% HCI in 190.0 ml 99% methanol and used as reagent. A rhodanine assay as described by Inoue and Hagerman (1988) was performed on the purified tannin crystals of twelve of the dietary plant species to determine the possible occurrence of hydrolysable tannin within the crystals. Five milligram crystals was added to 5.0 ml (2 N) HrSO,, the solution was then frozen, vacuum sealed and heated for 26 hr at 100°C. After hydrolysis the sample was opened and diluted to 50.0 ml with distilled water. In a test tube 1.5 ml methanolic rhodanine solution (0.667%) was added to 1.0 ml of the sample. After 5min, l.Oml (0.5 N) aqueous KOH was added. After a further 2.5 min the mixture was diluted to 25.0 ml with distilled water. Ten minutes later the absorbance was read at AA,, . A standard calibration curve was set using appropriate aliquots of a gallic acid standard made up to 1.Oml with (0.2 N) H,SO,. Of the leaves removed from the trees for analysis 1.00 g were immediately weighed in the field and added to 200.0 ml 99% methanol, and about 30g wet leaf material was weighed into a paper bag and later oven dried in order to determine the dry weight. Immediate fixation in 99% methanol prevents the gradual reduction in condensed tannin due to oxidation when stored in the dried form (McLeod, 1974; Price et ol., 1978). The leaves in methanol were put into a domestic liquid&r, milled for 3 min and left at room temperature in the dark for 4 days for total extraction. Of the supematant 1.Oml was added to 5.0 mI of the vanillin-HCI reagent and a further I.0 ml added to 5.0ml of the same reagent excluding the vanillin component (taken as blank; Price and Butler, 1977). Absorbance of both mixtures were measured at AA,s,, The absorbance of the blank was subtracted from that of the Vanillin containing mixture. The remaining absorbance value was read against the standard tannin curve of the plant species.
D.
FURSTENBURG
Four 2OO.Omg aliquots of the milled dried leaves were placed in fermentation bottles with 15.0 ml (0.1 M) phosphatt+bicarbonate buffer and 5.0 ml strained rumen fluid obtained from kudu inhabiting the same area in which the leaf samples of giraffe diet were collected. In vitro fermentation was done as described by Boomker and Van Hoven (1983) with the addition of 10~1 (0.05M) pivalic acid as internal standard for measuring vdlatile fatty acid (VFA) nroduction (Cxerkowski. 19761.Sub-samoles (0.5 ml) were do&ted at zero-time and ‘at hourly intervals, acidified to pH 1 and analysed using a Pye Unicam GCD chromatograph with a flame ionization detector and a glass column (4 mm diameter, 1000 mm length) packed with 60/80 Carbopack C/0.3% Carbowax 20 M/0.1% H,PQ as described by Dicorcia and Samperia (1974). Giraffes were followed by a 4 x 4 vehicle in the central district of the Kruger National Park, South Africa, over the period February 1985 to March 1986. The study area was approximate 2365 km* bushveld savanna of which 785 km2 was a Sclerocarya birrea-Acacia nigrescem open savanna. Approximately 3730 giraffes inhabited the area. Giraffe tracking was done under all weather conditions with varying following distance of 70 to 150 metres depending on the animals’ tolerance to the vehicle. The giraffe population was observed as a whole, no individuals were marked for permanent identification. In total 2730 observations (plants utilized) were made. From February 1986 to April 1987 browse samples of the observed giraffe diet were collected and analysed for condensed tannin content (% tannin/g/cm3 dried leaf material). RESULTS
A different calibration curve was obtained when using the tannin extract of each different plant species (Fig. 1) which points to a specificity of condensed tannins. The degree to which tannin concentration can be over or under estimated when using catechin as standard can be seen in Fig. 1. The light absorbance gradient of six plant species were repeated four times and gave identical slopes each time. Standard calibration curves that resulted from extracts with high light absorbance at AA, possibly might have tannins with high molecular weight as opposed to those with low light absorbance with molecular weight apparently -C500 (Goldstein and
Abafjrptbn at 490
nm
Fig. 1. Standard calibration lines obtained from catechin and purified tannins from the leaves of several tree species. Numbers correspond to the list in Table 1.
Condensed tannins Table
1.Giraffe dietary plants and condensed tannins as percentage of leaf dry matter per leaf volume Condensed tannin (%)
No.
species
Range
1
Acacia nilotica Peltophorum africanton Combretum reyheri Combretum apiculatum Euclea divinorum Terminalin prunioides Combretum hereroense Maytenus beterophylla Acacia nigrescens* Acacia robusta Acacia tort&s* Acacia welwitshii* Dichrostachys cbwrea* Larmea stuhlmanii Schotia brachypetalk Acacia exuvialis Ziziphus mucronata Combretum bnberbe’
6.61-27.67 1.70-18.08 0.81-6.07 3.67-38.20 4.01-26.62 0.02-12.92 I .99-34.19 1.65-28.17 0.01-10.23 0.22-2.72 0.01-5.94 0.1 a-4.97 0.16-10.64 3.64-49.3s 0.4320.43 0.35-10.76 0.01-1.82 0.031.50
2 3 4 5 6 I 8 9 10 11 12 13 14 I5 16 17 18
N Average samples 16.19 10.61 2.71 18.15 12.71 7.91 15.17 9.22 3.33 I .30 I .32 2.36 3.21 33.21 10.33 6.64 0.18 0.22
10 7 5 9 10 11 12 6 185 7 16 13 12 I 9 5 14 IO
Table 1 were all collected from the central district of the Kruger National Park. It is worth mentioning that none of the plants with more than 6% condensed tannin were of any dietary importance to giraffes within the Park. In terms of the influence of condensed tannin on metabolizable energy yield during kudu fermentation, only plant species with tannin levels lower than 20% were used in Fig. 2. These represent plants with possibly lower molecular weight tannins which have been shown by Goldstein and Swain (1965) to bind with proteins most effectively. The lowering of VFA energy yield with the increase in percentage tannin in the leaf could be the result of its binding and inhibition of microbial enzymes. DISCUSSION
‘These plants formed 83% of the diet of giraffe in the Kruger National Park.
Swain, 1963; Martin and Martin, 1982; Porter et al., 1986). None of the purified tannin crystals contained more than 10.6% hydrolysable tannin. Keeping in mind that the vanillin assay detects only condensed tannins (Martin and Martin, 1982; Hagerman and Butler, 1989; Mole et al., 1989) and some dihydrochalcones and anthocyanins (Sarkar and Howarth, 1976), the small possible appearance of hydrolysable tannin in the purified extract is of little concern. The other detectable non-tannin flavonoids have mostly been eliminated by the sephadex gel filtration. Using the standard calibration slopes obtained, extracts of these plants forming part of giraffe, kudu and impala diet within the study area, were tested for percentage tannin per unit leaf volume in the dry weight of the freshly sampled leaf (Table 1). The highest condensed tannin level was found in Lunnea stuhlmanii with 49.34% (Table 1). The plants listed in
0’
383
Proteins are amino acid polymers in the form of a helix due to, and kept in place by, hydrogen bonds. In order to precipitate protein the tannins must alter the hydrogen bonds responsible for the helix form with exposed -OH molecules on the flavanol units. Uptake of H,O forms part of the binding process of precipitation. Enough polymers of flavan-3-01 and or flavan-3,4-diol (condensed tannin) must be available to bind with the whole helix before actual precipitation takes place (Bate-Smith, 1973; Haslam, 1974; Asquith and Butler, 1985). On a quantitative level therefore, precipitation should be more effective if a plant has a moderate amount of low molecular weight condensed tannin than a high amount of high molecular weight condensed tannin. The large polymerized tannins do not dissolve under natural conditions and are too large to fit into the helixes in order to bind with hydrogen. Maximum precipitation occurs with molecules of between 300 and 700 molecular weight (Goldstein and Swain, 1963; Haslam and Lilley, 1986). Hagetman and Butler (1981) showed that some proteins have an extremely high affinity for tannin and that other proteins have a much lower affinity.
#
2
4
6
4
lb
ii
lh
16
18
20
Tannh % Fig. 2. The relationship between condensed tannin content of a plant and the amount of VFA energy it produces when fermented in o&o.
384
W. VANHOVENand D. FU~STI?NBURG
Those with a high affinity for tannin are all proline rich proteins. Proline, an imino acid, has a secondary amine nitrogen. Hagerman and Klucher (1986) shows that the carbonyl oxygen adjacent to a secondary amine nitrogen is a very good hydrogen bond acceptor, so proline rich proteins form especially strong hydrogen bonds with tannin. The presence of proline also restricts the protein molecular structure to some random coils as opposed to the regular firm helix formation, thus increasing the accessibility of the peptide backbone for hydrogen bonding. Hagerman and Klucher (1986) further shows that the interaction of proteins with tannin is very protein specific but is not tannin specific. Condensed tannin is slightly more effective as a protein precipitant than is hydrolysable tannin on a molar or weight basis. Asquith and Butler (1986) found that several different condensed tannins have high precipitation affinity for a certain prolinerich mouse salivary protein and no affinity for soybean tripsin enzyme. Austin et al. (1989) have shown that the saliva of a browser contains tannin-binding proteins whereas it is absent in grazers. Robbins et al. (1987b) also in studying mule deer as a browser and cows and domestic sheep as grazers, indicated that the browser has a proline rich protein in the saliva which is not the case with the grazers. These findings suggest that the tolerance of browsers for dietary tannins may be due to the production of salivary proline rich proteins. Minimal biological significance can be attached to condensed tannin estimation if the tannin produced by the particular plant is not first isolated to determine its optical density which is related to its molecular weight. Porter et al. (1986) and Wisdom et al. (1987) show the different activities of different condensed tannins with the n-BuOH-HCl assay and critic&s the use of a single tannin standard for estimating tannin content of plant samples of different tannin flavanol composition. Tempel ( 1982) notes the ignorance of species variability of tannin concentrations in ecological literature which can be referred to the use of single standards in tannin assays. Hagerman and Butler (1989) are in favour of the use of purified plant tannin extracts as standards for comparisons between different plant species. It is evident from Fig. 1 that condensed tannin in terms of its optical density is largely species specific for plants. This in turn has a bearing on its binding potential to protein and enzymes and thus its ability to interfere with digestion. One of the most important functions of fermentation in the rumen is the production of metabolizable energy in the form of volatile fatty acids which are end products of carbohydrate fermentation. In order to interpret the potential inhibiting effect of tannin on fermentation energy yield, Fig. 2 illustrates the rapid decline in VFA energy yield by fermentation as the tannin values in the substrate increases to and above the 6% level. The influence of relatively high levels of tannin in the diet on the fermentation process and metabolizable energy yield, should in the live browser be less than is shown in Fig. 2. These dietary tannins used in the in vitro fermentation have not been exposed to any saliva prior to entering the microbial fermentation environment. This does emphasize the
importance of a salivary tannin binding agent to dilute the negative influence which tannins could have in the rumen. Theoretically an animal should eat more of a plant with tannin levels higher than 6% in order to comply with its energy demands. In practice however, it rather avoids those plants (Cooper and Owen-Smith, 1985). Our own observations over 12 months in the central district of the Kruger National Park proved that of 23 plant species recorded to have been browsed by giraffes, five formed 83% of their total diet. These five plant species had the lowest tannin values, less than 6% (Table 1). The more accurate condensed tannin assay used confirms the avoidance by African ruminant browers of plants with higher than 6% condensed tannin. If they were forced to consume a diet of higher tannin value, the tannin could interfere with the fermentation process to such an extent that available metaboliible energy from this source would become limiting. REFERENCES
Arnold G. W. and Hill J. L. (1972) Chemical factors affecting selection of food plants by ruminants. In Phytochemicaf Ecology (Edited by Harbome J. B.), pp. 71-101. Academic Press, London. Asquith T. N. and Butler L. G. (1985) Use of dye-labeled protein as spectrophotometric assay for protein precipitants such as tannin. J. C/rem. Ecol. 11, 1525-1544. Aaquith T. N. and Butler L. G. (1986) Interactions of condensed tannins with selected proteins. Phytochemistry zs, 1591-1593. Austin P. J., Suchar L. A., Robbins C. T. and Hagerman Ann E. (1989) Tannin-binding proteins in saliva of deer and their absence in saliva of sheep and cattle. J. Chem. Ecol. 15, 1335-1347. Bate-Smith E. C. (1973) Haemonolysis of tannins: tire concept of relative astringency. Phytochembtry 12, 907-912. Beart J. E., Lilley T. H. and Haslam E. (1985) Plantpolyphenols secondary metabolism and chemical defense: some observations. Phytochembrry 24, 33-38. Boomker Elizabeth A. and Van Hoven W. (1983) In oitro digestibility of plants normally consumed by the kudu, Tragelaphus strepsiceros. S. Afr. J. Anim. Sci. 13, 206-207. Bums R. E. (1971) Method for estimation of tannin in grain sorghum. Agronomy J. 63, 511-512. Cooper S. M. and Owen-Smith N. (1985) Condensed tannins deter feeding by browsing ruminants in a South African savanna. Oecologia 67, 142-146. Cooper S. M., Owen-Smith N. and Bryant J. P. (1988) Foliage acceptability to browsing ruminants in relation to seasonal changes in the leaf chemistry of woody plants in a South African savanna. Oecologia75, 336-342. Czerkowski J. W. (1976) The use of pivalic acid as a reference substance in measurements of production of volatile fatty acids by rumen microorganisms ia vitro. Er. J. Nutr. 36, 311-321. Dicorcia A. and Samperia R. (1974) Determination of trace amounts of C2C5 acids in aqueous solutions by gas chromatography. Anaiyt. Chem. 46, 140-143. Feath S. H. and Bultman T. L. (1986) Interacting effects of increased tannin levels on leaf-mining insects. l3tomol. Exp. Appl. 40, 297-300. Feeny P. P. (1976) Plant apparency and chemical defense. Rec. A&. Phytochem. 10, 140.
Condensed tannins Freeland W. J. and Janzen D. H. (1974) Strategies in herbivory by mammals: The role of plant secondary compounds. Am. No. 108, 269-289. Geissmann T. A. and Hinreiner E. (1952) Theories of the biogenesis of flavonoid compounds (Part 1). Sot. Rev. 18, 77-163. Goldstein J. L. and Swain T. (1963) Changes in tannins in ripening of fruits. Phyrochemisrry 2, 371-383. Goldstein J. L. and Swain T. (1965) The inhibition of enzymes by tannins. Phyfochemistry 4, 185-192. Hagerman Ann E. and Butler L. G. (1978) Protein precipitation method for the quantitative determination of tannins. J. Agric. Fd. Chem. 26, 809-812. Hagerman Ann E. and Butler L. G. (1981) The specificity of proanthocyanidin protein interactions. J. biol. Chem. 256,4494-4497.
Hagerman Ann E. and Butler L. G. (1989) Choosing appropriate methods and standards for assaying tannin. J. Chem. Ecol. 11, 1535-1544. Hagerman Ann E. and Klucher K. M. (1986) Tannin-protein interactions. In Plant Flavonoids in Biology and Medicine : Biochemical, Pharmacological and Structure Activity Relationships (Edited by Cody V., Middleton E.
and Harbome J.), pp. 67-76. Alan R. Liss, Inc., New York. Harbome J. B. (1967) Comparative Biochemistry of Flavonoids. Academic Press, London. Haslam E. (1974) Polyphenol-protein interactions. Biochem. J. 139, 285-288. Haslam E. (1977) Symmetry and promiscuity in procyanidin biochemistry. Phytochemistry 16, 1625-1640. Haslam E. and Lilley T. H. (1986) Interactions of natural phenols with macromolecules. In Plant Flavonoids in Biology and Medicine : Biochemical, Pharmacological and Structure Activity Relationships (Edited by Cody V.,
Middleton E. and Harbome J.), pp. 53-65. Alan R. Liss, Inc., New York. Haukioja E. (1980) On the role of plant defences in the fluctuation of herbivore populations. Oikos 35, 202-213. Inoue K. H. and Hagennan Ann E. (1988) Determination of gallotannin with rhodanine. Analyt. Biochem. 169, 363-369. Kearsley M. J. C. and Whitham T. G. (1989) Developmental changes in resistance to herbivory: implications for individuals and populations. Ecology 70, 422-434. Martin J. S. and Martin M. M. (1982) Tannin assays in ecological studies: lack of correlation between phenolics, proanthocyanidins, and protein-precipitating constituents in mature foliage of six oak species. Oecologia 54, 20-211.
McLeod M. N. (1974) Plant tannins and their role in forage quality. Commonwealth Bureau of Nutrition. Nutr. Abs. Rev. 44, 803-815.
Mole S., Butler L. G., Hagerman Ann E. and Waterman P. (1989) Ecological tannin assays: a critique. Oecologia 78, 93-96.
Porter L. J., Hrstich L. N. and Chan B. G. (1986) The conversion of prodelphinidins to cyanidin and delphinidin. Phytochemistry 25, 223-230.
CBP(A)
101:2-N
385
Price M. L. and Butler L. G. (1977) Rapid visual estimation and spectrophotometric determination of tannin content of sorghum grain. J. Agric. Fd. Chem. 25, 1268-1273. Price M. L., Van Scoyoc S. and Butler L. G. (1978) A critical evaluation of the vanillin reaction as an assay for tannin in sorghum grain. J. Agric. Fd. Chem. 26, 12141218. Repas A., Nikolin B. and Dursun K. (1969) The separation of phenolic glucosides by gel filtration. J. Chromatog. 44, 184-187.
Rhoades D. F. and Cates R. G. (1976) Toward a general theory of plant antiherbivore chemistry. Rec. Adv. Phytochem. 10, 168-213. Rhoades D. F. and Wooltorton L. S. C. (1978) The biosynthesis of phenolic compounds in wounded plant storage tissues. In Biochemistry of Wounded Plant Ttkues (Edited by Kahl E. G.), pp. 243-286. Walter de Gruyter, Berlin. Rhoades D. F. (1983) Responses of alder and willow to attack by tent caterpillars and webworms: evidence for pheromdnal sensitivity of willows. Plant resistance to insects: A.G.S. Svmvosium Ser. u)8. 55-68. Rhoades D. F. (1985) Gffensivedefensive interactions between herbivores and plants: their relevance in herbivore population dynamics and ecological theory. Am. Nat. lU, 205-238. Robbins C. T., Hanley T. A., Hagerman Ann E., Hjeljord 0.. Baker D. L.. Schwart C. C. and Mautz W. W. (1987a) Role of tannins in defending plants against ruminants: reduction in protein availabihty. Ecology 68, 98-107. Robbins C. T.. Mole S.. Haaerman Ann E. and Hanlev T. A. (1987b) Role of t&n& in defending plants against ruminants: reduction in dry matter digestion. Ecology 68, 16061615. Sarkar S. K., Howarth R. E. and Goplen B. P. (1976) Condensed tannins in herbaceous legumes. Crop. Sci. 16, 543-546. Sarkar S. K. and Howarth R. E. (1976) Specificity of the vanillin test for flavanols. J. Agric. Fd. Chem. 24,3 17-320. Strumeyer D. H. and Malin M. J. (1975) Condensed tannins in grain sorghum: isolation, fractionation and characterization. J. Agric. Fd. Chem. 13, 909-914. Swain T. (1977) Secondary compounds as protective agents. A. Rev. Plant Phys. 28, 479-501.
Tempel A. S. (1982) Tannin-measuring techniques: a review. J. Chem. Ecol. 8, 1289-1299.
TouzC A. and Esquerre-Tugaye M. (1980) Defence mechanisms of plants against varietal non-specific pathogens. In Active Defence Mechanisms in Plants (Edited bv Wood R. K. S.); pp. 103-l 17. Plenum Press, ‘New York. Waring G. L. and Price P. W. (1988) Consequences of host plant chemical and physical variability to an associated herbivore. Ecol. Res. 3, 205-216. Wisdom C. S., Gonazalez-Coloma A. and Rundel P. W. (1987) Ecological tannin assays: evaluation of proanthocyanidins, protein binding assays and protein preciptating potential. Oecologia 72, 395-401. Zucker W. V. (1983) Tannins: does structure determine function? An ecological perspective. Am. Nat. 121, 335-365.
No.
0300-9629/92$5.00+ 0.00 0 1992Pergamon Press plc
2, pp. 381-385,1992
Printed in Great Britain
THE USE OF PURIFIED CONDENSED TANNIN AS A REFERENCE IN DETERMINING ITS INFLUENCE ON RUMEN FERMENTATION W.
VAN
HOVEN* and D. FUR~TENBURG
Centre for Wildlife Research, University of Pretoria, Pretoria 0002, South Africa. Telephone: (0 12) 420-2627 [Received 29 April 1991) Ah&net-I. Tannins were purified from the leaves of trees forming part of giraffe’s diet in the Kruger National Park. 2. In general, hydrolysable tannin formed less than 10% of the total purified crystal complexes, condensed tannin and possibly some non tannin phenols forming the balance. 3. Bach tree species condensed tannin gave a different callibration curve and these were used in the assay. 4. Volatile fatty acid production during in vitro fermentation was greatly reduced when the substrate contained more than 6% condensed tannin.
INTRODIJCI’ION
The role of tannins as a chemical deterent to herbivory has been well documented (Arnold and Hill, 1972; Freeland and Janzen, 1974; McLeod, 1974; Feeny, 1976; Rhoades and Cates, 1976; Swain, 1977; Haukioja, 1980; Touze and Esquerre-Tugaye, 1980; Rhoades, 1983,1985; Zucker, 1983; Beart et al., 1985; Cooper and Gwen-Smith, 1985; Cooper et al., 1988; Feath and Bultman, 1986; Robbins et al., 1987a,b; Waring and Price, 1988; Kearsley and Whitham, 1989). The term tannin is a general one for a large variety of combinations of basic phenolic (C,) molecules. The flavonoids are C&& molecules and their derivatives. The two major types of tannin (condensed and hydrolysable) are chemically quite different. Condensed tannins are divided into (A) proanthocyanidins which are di-, tri- or polymers of the anthocyanidin and or catechin flavan-3-01s and (B) leucoanthocyanidins which are dimers of the flavan3,4-diol flavonoids (Geissman and Hinreiner, 1952; Harborne, 1967; Sarkar et al., 1976; Rhoades and Wooltorton, 1978; Hagerman and Butler, 1989). Catechin is probably the best known due to its wide use as a standard in the vanillin assay of tannin. Hydrolysable tannins are gallic or hexahydroxydiphenic acid esters of glucose or other polyols (Haslam, 1977). The binding of proteins to tannins varies according to the types of proteins and tannins. The assays presently available measuring protein for binding/precipitation are not satisfactory in all respects (Asquith and Butler, 1985; Hagerman and Butler, 1989; Mole et al., 1989). Asquith and Butler (1985) showed that five different proteins varied by as much as three orders of magnitude in their relative affinities for sorghum grain tannin. Thus, owing to the large variety of anthocyanidins and condensed *To whom correspondence
should be addressed.
tannins that occur in plants, the use of catechin or any other single purified tannin as a standard in calibration could lead to inaccuarate estimates as shown by Wisdom et al. (1987). In this study a colometric assay based on the original description of Bums (1971) was used. Tannin
extracts were made of each plant to be analysed and the purified tannin thus obtained was used to set up a calibration curve for each particular plant species. When using catechin for a standard calibration curve, tannin values in grain sorghum were over estimated by two-fold at AA,, for absorbance values lower than 0.2 and by 3-fold for values higher than 1.4 (Hagerman and Butler, 1978; Price et al., 1978). In studying the influence which tannins have on the feeding, digestion and ecology of browsing ruminants, a reliable practical analytical procedure had to be established. Purifying a particular plant species tannin and using it to set up a calibration curve is advantageous in that this curve can be used for all further analyses of that particular plant species, provided the same asssay is used. Results obtained by means of this analytical procedure were related to the volatile fatty acid energy that becomes available as metabolizable energy during fermentation. MATERIALS
AND
METHODS
Based on the method of Strumeyer and Malin (1975) a 2.0 x 28.0 cm column was packed with 12 g Sephadex LH20 gel saturated with 99% methanol. Leaves of trees that occur in the diet of giraffe were collected, dried and ground in a laboratory mill. A log sample was washed four times in 60ml of peroxide-free ether by shaking the flasks at regular intervals over 3 hr and finally centrifuging, thus effectively removing the fats, fatty acids, essential oils and most chlorophyll. The sediment containing the flavonoids was resuspended in 200 ml 99% methanol-and left air sealed in the dark at room temperature for 48 hr. This 200 ml flavonol extract was filtered throuah the gel column and rinsed with an additional 80ml of&e
381
382
W.
VAN HOVEN and
methanol, thus removing most of the non-tannin flavonoids from the gel (Renas ef al.. 1969; Strumever and Malin. 1975). The-tannins precipitated to the gel were removed by rinsing the column with 70 ml eluent of 50% aqueous acetone. The filtrate was evaporated to the remain of dark rust-brown tannin crystals. Crystalixed tannin was immediately redissolved in 99% methanol in a ratio of 75.0 mg: 10.0 ml and stored air sealed at -4°C in the dark until the calibration dilutions were made based on Bums (1971). The dilution series started with a concentration of 1.125 mg tannin per ml methanol. Vanillin (2.500 g) in 250.0 ml 99% methanol was added to 60.0 ml 32% HCI in 190.0 ml 99% methanol and used as reagent. A rhodanine assay as described by Inoue and Hagerman (1988) was performed on the purified tannin crystals of twelve of the dietary plant species to determine the possible occurrence of hydrolysable tannin within the crystals. Five milligram crystals was added to 5.0 ml (2 N) HrSO,, the solution was then frozen, vacuum sealed and heated for 26 hr at 100°C. After hydrolysis the sample was opened and diluted to 50.0 ml with distilled water. In a test tube 1.5 ml methanolic rhodanine solution (0.667%) was added to 1.0 ml of the sample. After 5min, l.Oml (0.5 N) aqueous KOH was added. After a further 2.5 min the mixture was diluted to 25.0 ml with distilled water. Ten minutes later the absorbance was read at AA,, . A standard calibration curve was set using appropriate aliquots of a gallic acid standard made up to 1.Oml with (0.2 N) H,SO,. Of the leaves removed from the trees for analysis 1.00 g were immediately weighed in the field and added to 200.0 ml 99% methanol, and about 30g wet leaf material was weighed into a paper bag and later oven dried in order to determine the dry weight. Immediate fixation in 99% methanol prevents the gradual reduction in condensed tannin due to oxidation when stored in the dried form (McLeod, 1974; Price et ol., 1978). The leaves in methanol were put into a domestic liquid&r, milled for 3 min and left at room temperature in the dark for 4 days for total extraction. Of the supematant 1.Oml was added to 5.0 mI of the vanillin-HCI reagent and a further I.0 ml added to 5.0ml of the same reagent excluding the vanillin component (taken as blank; Price and Butler, 1977). Absorbance of both mixtures were measured at AA,s,, The absorbance of the blank was subtracted from that of the Vanillin containing mixture. The remaining absorbance value was read against the standard tannin curve of the plant species.
D.
FURSTENBURG
Four 2OO.Omg aliquots of the milled dried leaves were placed in fermentation bottles with 15.0 ml (0.1 M) phosphatt+bicarbonate buffer and 5.0 ml strained rumen fluid obtained from kudu inhabiting the same area in which the leaf samples of giraffe diet were collected. In vitro fermentation was done as described by Boomker and Van Hoven (1983) with the addition of 10~1 (0.05M) pivalic acid as internal standard for measuring vdlatile fatty acid (VFA) nroduction (Cxerkowski. 19761.Sub-samoles (0.5 ml) were do&ted at zero-time and ‘at hourly intervals, acidified to pH 1 and analysed using a Pye Unicam GCD chromatograph with a flame ionization detector and a glass column (4 mm diameter, 1000 mm length) packed with 60/80 Carbopack C/0.3% Carbowax 20 M/0.1% H,PQ as described by Dicorcia and Samperia (1974). Giraffes were followed by a 4 x 4 vehicle in the central district of the Kruger National Park, South Africa, over the period February 1985 to March 1986. The study area was approximate 2365 km* bushveld savanna of which 785 km2 was a Sclerocarya birrea-Acacia nigrescem open savanna. Approximately 3730 giraffes inhabited the area. Giraffe tracking was done under all weather conditions with varying following distance of 70 to 150 metres depending on the animals’ tolerance to the vehicle. The giraffe population was observed as a whole, no individuals were marked for permanent identification. In total 2730 observations (plants utilized) were made. From February 1986 to April 1987 browse samples of the observed giraffe diet were collected and analysed for condensed tannin content (% tannin/g/cm3 dried leaf material). RESULTS
A different calibration curve was obtained when using the tannin extract of each different plant species (Fig. 1) which points to a specificity of condensed tannins. The degree to which tannin concentration can be over or under estimated when using catechin as standard can be seen in Fig. 1. The light absorbance gradient of six plant species were repeated four times and gave identical slopes each time. Standard calibration curves that resulted from extracts with high light absorbance at AA, possibly might have tannins with high molecular weight as opposed to those with low light absorbance with molecular weight apparently -C500 (Goldstein and
Abafjrptbn at 490
nm
Fig. 1. Standard calibration lines obtained from catechin and purified tannins from the leaves of several tree species. Numbers correspond to the list in Table 1.
Condensed tannins Table
1.Giraffe dietary plants and condensed tannins as percentage of leaf dry matter per leaf volume Condensed tannin (%)
No.
species
Range
1
Acacia nilotica Peltophorum africanton Combretum reyheri Combretum apiculatum Euclea divinorum Terminalin prunioides Combretum hereroense Maytenus beterophylla Acacia nigrescens* Acacia robusta Acacia tort&s* Acacia welwitshii* Dichrostachys cbwrea* Larmea stuhlmanii Schotia brachypetalk Acacia exuvialis Ziziphus mucronata Combretum bnberbe’
6.61-27.67 1.70-18.08 0.81-6.07 3.67-38.20 4.01-26.62 0.02-12.92 I .99-34.19 1.65-28.17 0.01-10.23 0.22-2.72 0.01-5.94 0.1 a-4.97 0.16-10.64 3.64-49.3s 0.4320.43 0.35-10.76 0.01-1.82 0.031.50
2 3 4 5 6 I 8 9 10 11 12 13 14 I5 16 17 18
N Average samples 16.19 10.61 2.71 18.15 12.71 7.91 15.17 9.22 3.33 I .30 I .32 2.36 3.21 33.21 10.33 6.64 0.18 0.22
10 7 5 9 10 11 12 6 185 7 16 13 12 I 9 5 14 IO
Table 1 were all collected from the central district of the Kruger National Park. It is worth mentioning that none of the plants with more than 6% condensed tannin were of any dietary importance to giraffes within the Park. In terms of the influence of condensed tannin on metabolizable energy yield during kudu fermentation, only plant species with tannin levels lower than 20% were used in Fig. 2. These represent plants with possibly lower molecular weight tannins which have been shown by Goldstein and Swain (1965) to bind with proteins most effectively. The lowering of VFA energy yield with the increase in percentage tannin in the leaf could be the result of its binding and inhibition of microbial enzymes. DISCUSSION
‘These plants formed 83% of the diet of giraffe in the Kruger National Park.
Swain, 1963; Martin and Martin, 1982; Porter et al., 1986). None of the purified tannin crystals contained more than 10.6% hydrolysable tannin. Keeping in mind that the vanillin assay detects only condensed tannins (Martin and Martin, 1982; Hagerman and Butler, 1989; Mole et al., 1989) and some dihydrochalcones and anthocyanins (Sarkar and Howarth, 1976), the small possible appearance of hydrolysable tannin in the purified extract is of little concern. The other detectable non-tannin flavonoids have mostly been eliminated by the sephadex gel filtration. Using the standard calibration slopes obtained, extracts of these plants forming part of giraffe, kudu and impala diet within the study area, were tested for percentage tannin per unit leaf volume in the dry weight of the freshly sampled leaf (Table 1). The highest condensed tannin level was found in Lunnea stuhlmanii with 49.34% (Table 1). The plants listed in
0’
383
Proteins are amino acid polymers in the form of a helix due to, and kept in place by, hydrogen bonds. In order to precipitate protein the tannins must alter the hydrogen bonds responsible for the helix form with exposed -OH molecules on the flavanol units. Uptake of H,O forms part of the binding process of precipitation. Enough polymers of flavan-3-01 and or flavan-3,4-diol (condensed tannin) must be available to bind with the whole helix before actual precipitation takes place (Bate-Smith, 1973; Haslam, 1974; Asquith and Butler, 1985). On a quantitative level therefore, precipitation should be more effective if a plant has a moderate amount of low molecular weight condensed tannin than a high amount of high molecular weight condensed tannin. The large polymerized tannins do not dissolve under natural conditions and are too large to fit into the helixes in order to bind with hydrogen. Maximum precipitation occurs with molecules of between 300 and 700 molecular weight (Goldstein and Swain, 1963; Haslam and Lilley, 1986). Hagetman and Butler (1981) showed that some proteins have an extremely high affinity for tannin and that other proteins have a much lower affinity.
#
2
4
6
4
lb
ii
lh
16
18
20
Tannh % Fig. 2. The relationship between condensed tannin content of a plant and the amount of VFA energy it produces when fermented in o&o.
384
W. VANHOVENand D. FU~STI?NBURG
Those with a high affinity for tannin are all proline rich proteins. Proline, an imino acid, has a secondary amine nitrogen. Hagerman and Klucher (1986) shows that the carbonyl oxygen adjacent to a secondary amine nitrogen is a very good hydrogen bond acceptor, so proline rich proteins form especially strong hydrogen bonds with tannin. The presence of proline also restricts the protein molecular structure to some random coils as opposed to the regular firm helix formation, thus increasing the accessibility of the peptide backbone for hydrogen bonding. Hagerman and Klucher (1986) further shows that the interaction of proteins with tannin is very protein specific but is not tannin specific. Condensed tannin is slightly more effective as a protein precipitant than is hydrolysable tannin on a molar or weight basis. Asquith and Butler (1986) found that several different condensed tannins have high precipitation affinity for a certain prolinerich mouse salivary protein and no affinity for soybean tripsin enzyme. Austin et al. (1989) have shown that the saliva of a browser contains tannin-binding proteins whereas it is absent in grazers. Robbins et al. (1987b) also in studying mule deer as a browser and cows and domestic sheep as grazers, indicated that the browser has a proline rich protein in the saliva which is not the case with the grazers. These findings suggest that the tolerance of browsers for dietary tannins may be due to the production of salivary proline rich proteins. Minimal biological significance can be attached to condensed tannin estimation if the tannin produced by the particular plant is not first isolated to determine its optical density which is related to its molecular weight. Porter et al. (1986) and Wisdom et al. (1987) show the different activities of different condensed tannins with the n-BuOH-HCl assay and critic&s the use of a single tannin standard for estimating tannin content of plant samples of different tannin flavanol composition. Tempel ( 1982) notes the ignorance of species variability of tannin concentrations in ecological literature which can be referred to the use of single standards in tannin assays. Hagerman and Butler (1989) are in favour of the use of purified plant tannin extracts as standards for comparisons between different plant species. It is evident from Fig. 1 that condensed tannin in terms of its optical density is largely species specific for plants. This in turn has a bearing on its binding potential to protein and enzymes and thus its ability to interfere with digestion. One of the most important functions of fermentation in the rumen is the production of metabolizable energy in the form of volatile fatty acids which are end products of carbohydrate fermentation. In order to interpret the potential inhibiting effect of tannin on fermentation energy yield, Fig. 2 illustrates the rapid decline in VFA energy yield by fermentation as the tannin values in the substrate increases to and above the 6% level. The influence of relatively high levels of tannin in the diet on the fermentation process and metabolizable energy yield, should in the live browser be less than is shown in Fig. 2. These dietary tannins used in the in vitro fermentation have not been exposed to any saliva prior to entering the microbial fermentation environment. This does emphasize the
importance of a salivary tannin binding agent to dilute the negative influence which tannins could have in the rumen. Theoretically an animal should eat more of a plant with tannin levels higher than 6% in order to comply with its energy demands. In practice however, it rather avoids those plants (Cooper and Owen-Smith, 1985). Our own observations over 12 months in the central district of the Kruger National Park proved that of 23 plant species recorded to have been browsed by giraffes, five formed 83% of their total diet. These five plant species had the lowest tannin values, less than 6% (Table 1). The more accurate condensed tannin assay used confirms the avoidance by African ruminant browers of plants with higher than 6% condensed tannin. If they were forced to consume a diet of higher tannin value, the tannin could interfere with the fermentation process to such an extent that available metaboliible energy from this source would become limiting. REFERENCES
Arnold G. W. and Hill J. L. (1972) Chemical factors affecting selection of food plants by ruminants. In Phytochemicaf Ecology (Edited by Harbome J. B.), pp. 71-101. Academic Press, London. Asquith T. N. and Butler L. G. (1985) Use of dye-labeled protein as spectrophotometric assay for protein precipitants such as tannin. J. C/rem. Ecol. 11, 1525-1544. Aaquith T. N. and Butler L. G. (1986) Interactions of condensed tannins with selected proteins. Phytochemistry zs, 1591-1593. Austin P. J., Suchar L. A., Robbins C. T. and Hagerman Ann E. (1989) Tannin-binding proteins in saliva of deer and their absence in saliva of sheep and cattle. J. Chem. Ecol. 15, 1335-1347. Bate-Smith E. C. (1973) Haemonolysis of tannins: tire concept of relative astringency. Phytochembtry 12, 907-912. Beart J. E., Lilley T. H. and Haslam E. (1985) Plantpolyphenols secondary metabolism and chemical defense: some observations. Phytochembrry 24, 33-38. Boomker Elizabeth A. and Van Hoven W. (1983) In oitro digestibility of plants normally consumed by the kudu, Tragelaphus strepsiceros. S. Afr. J. Anim. Sci. 13, 206-207. Bums R. E. (1971) Method for estimation of tannin in grain sorghum. Agronomy J. 63, 511-512. Cooper S. M. and Owen-Smith N. (1985) Condensed tannins deter feeding by browsing ruminants in a South African savanna. Oecologia 67, 142-146. Cooper S. M., Owen-Smith N. and Bryant J. P. (1988) Foliage acceptability to browsing ruminants in relation to seasonal changes in the leaf chemistry of woody plants in a South African savanna. Oecologia75, 336-342. Czerkowski J. W. (1976) The use of pivalic acid as a reference substance in measurements of production of volatile fatty acids by rumen microorganisms ia vitro. Er. J. Nutr. 36, 311-321. Dicorcia A. and Samperia R. (1974) Determination of trace amounts of C2C5 acids in aqueous solutions by gas chromatography. Anaiyt. Chem. 46, 140-143. Feath S. H. and Bultman T. L. (1986) Interacting effects of increased tannin levels on leaf-mining insects. l3tomol. Exp. Appl. 40, 297-300. Feeny P. P. (1976) Plant apparency and chemical defense. Rec. A&. Phytochem. 10, 140.
Condensed tannins Freeland W. J. and Janzen D. H. (1974) Strategies in herbivory by mammals: The role of plant secondary compounds. Am. No. 108, 269-289. Geissmann T. A. and Hinreiner E. (1952) Theories of the biogenesis of flavonoid compounds (Part 1). Sot. Rev. 18, 77-163. Goldstein J. L. and Swain T. (1963) Changes in tannins in ripening of fruits. Phyrochemisrry 2, 371-383. Goldstein J. L. and Swain T. (1965) The inhibition of enzymes by tannins. Phyfochemistry 4, 185-192. Hagerman Ann E. and Butler L. G. (1978) Protein precipitation method for the quantitative determination of tannins. J. Agric. Fd. Chem. 26, 809-812. Hagerman Ann E. and Butler L. G. (1981) The specificity of proanthocyanidin protein interactions. J. biol. Chem. 256,4494-4497.
Hagerman Ann E. and Butler L. G. (1989) Choosing appropriate methods and standards for assaying tannin. J. Chem. Ecol. 11, 1535-1544. Hagerman Ann E. and Klucher K. M. (1986) Tannin-protein interactions. In Plant Flavonoids in Biology and Medicine : Biochemical, Pharmacological and Structure Activity Relationships (Edited by Cody V., Middleton E.
and Harbome J.), pp. 67-76. Alan R. Liss, Inc., New York. Harbome J. B. (1967) Comparative Biochemistry of Flavonoids. Academic Press, London. Haslam E. (1974) Polyphenol-protein interactions. Biochem. J. 139, 285-288. Haslam E. (1977) Symmetry and promiscuity in procyanidin biochemistry. Phytochemistry 16, 1625-1640. Haslam E. and Lilley T. H. (1986) Interactions of natural phenols with macromolecules. In Plant Flavonoids in Biology and Medicine : Biochemical, Pharmacological and Structure Activity Relationships (Edited by Cody V.,
Middleton E. and Harbome J.), pp. 53-65. Alan R. Liss, Inc., New York. Haukioja E. (1980) On the role of plant defences in the fluctuation of herbivore populations. Oikos 35, 202-213. Inoue K. H. and Hagennan Ann E. (1988) Determination of gallotannin with rhodanine. Analyt. Biochem. 169, 363-369. Kearsley M. J. C. and Whitham T. G. (1989) Developmental changes in resistance to herbivory: implications for individuals and populations. Ecology 70, 422-434. Martin J. S. and Martin M. M. (1982) Tannin assays in ecological studies: lack of correlation between phenolics, proanthocyanidins, and protein-precipitating constituents in mature foliage of six oak species. Oecologia 54, 20-211.
McLeod M. N. (1974) Plant tannins and their role in forage quality. Commonwealth Bureau of Nutrition. Nutr. Abs. Rev. 44, 803-815.
Mole S., Butler L. G., Hagerman Ann E. and Waterman P. (1989) Ecological tannin assays: a critique. Oecologia 78, 93-96.
Porter L. J., Hrstich L. N. and Chan B. G. (1986) The conversion of prodelphinidins to cyanidin and delphinidin. Phytochemistry 25, 223-230.
CBP(A)
101:2-N
385
Price M. L. and Butler L. G. (1977) Rapid visual estimation and spectrophotometric determination of tannin content of sorghum grain. J. Agric. Fd. Chem. 25, 1268-1273. Price M. L., Van Scoyoc S. and Butler L. G. (1978) A critical evaluation of the vanillin reaction as an assay for tannin in sorghum grain. J. Agric. Fd. Chem. 26, 12141218. Repas A., Nikolin B. and Dursun K. (1969) The separation of phenolic glucosides by gel filtration. J. Chromatog. 44, 184-187.
Rhoades D. F. and Cates R. G. (1976) Toward a general theory of plant antiherbivore chemistry. Rec. Adv. Phytochem. 10, 168-213. Rhoades D. F. and Wooltorton L. S. C. (1978) The biosynthesis of phenolic compounds in wounded plant storage tissues. In Biochemistry of Wounded Plant Ttkues (Edited by Kahl E. G.), pp. 243-286. Walter de Gruyter, Berlin. Rhoades D. F. (1983) Responses of alder and willow to attack by tent caterpillars and webworms: evidence for pheromdnal sensitivity of willows. Plant resistance to insects: A.G.S. Svmvosium Ser. u)8. 55-68. Rhoades D. F. (1985) Gffensivedefensive interactions between herbivores and plants: their relevance in herbivore population dynamics and ecological theory. Am. Nat. lU, 205-238. Robbins C. T., Hanley T. A., Hagerman Ann E., Hjeljord 0.. Baker D. L.. Schwart C. C. and Mautz W. W. (1987a) Role of tannins in defending plants against ruminants: reduction in protein availabihty. Ecology 68, 98-107. Robbins C. T.. Mole S.. Haaerman Ann E. and Hanlev T. A. (1987b) Role of t&n& in defending plants against ruminants: reduction in dry matter digestion. Ecology 68, 16061615. Sarkar S. K., Howarth R. E. and Goplen B. P. (1976) Condensed tannins in herbaceous legumes. Crop. Sci. 16, 543-546. Sarkar S. K. and Howarth R. E. (1976) Specificity of the vanillin test for flavanols. J. Agric. Fd. Chem. 24,3 17-320. Strumeyer D. H. and Malin M. J. (1975) Condensed tannins in grain sorghum: isolation, fractionation and characterization. J. Agric. Fd. Chem. 13, 909-914. Swain T. (1977) Secondary compounds as protective agents. A. Rev. Plant Phys. 28, 479-501.
Tempel A. S. (1982) Tannin-measuring techniques: a review. J. Chem. Ecol. 8, 1289-1299.
TouzC A. and Esquerre-Tugaye M. (1980) Defence mechanisms of plants against varietal non-specific pathogens. In Active Defence Mechanisms in Plants (Edited bv Wood R. K. S.); pp. 103-l 17. Plenum Press, ‘New York. Waring G. L. and Price P. W. (1988) Consequences of host plant chemical and physical variability to an associated herbivore. Ecol. Res. 3, 205-216. Wisdom C. S., Gonazalez-Coloma A. and Rundel P. W. (1987) Ecological tannin assays: evaluation of proanthocyanidins, protein binding assays and protein preciptating potential. Oecologia 72, 395-401. Zucker W. V. (1983) Tannins: does structure determine function? An ecological perspective. Am. Nat. 121, 335-365.
