Fluoride and metals in Byrsonima crassifolia,

a medicinal tree from the neotropical savannahs

Elizabeth Olivares and Eder Peña

Elizabeth Olivares. Ph.D. in Biology (Ecology), Instituto Venezolano de Investigaciones Científicas (IVIC). Researcher, IVIC. Address: Centro de Ecología, IVIC. Apartado 21827. Caracas 1020A. Venezuela. e-mail: eolivare@ivic.ve

Eder Peña. Bachelor in Biology, Universidad Central de Venezuela. Researcher Professional, IVIC. e-mail: epena@ivic.ve

Resumen

La concentración de fluoruro (F) en relación a elementos minerales (K, Ca, Mg, Mn, Fe, Cu, Zn, Ni, Co, Cr, Al, Pb y Cd) y fenoles totales fue evaluada en hojas de Byrsonima crassifolia. Se cosecharon hojas jóvenes, adultas y caídas en un jardín situado en las afueras de una industria de aluminio, en el Estado Bolívar, Venezuela y en dos sitios en la sabana a una distancia aproximada de 5 y 8km de la fábrica en la dirección del viento. Se encontró una regresión lineal positiva de F con Mn, Fe, Pb, Al, Ca, Mg, Cr y Zn para hojas jóvenes y adultas. Se colectaron también hojas en otros dos sitios más distantes (B1 y B2) y en un sitio contrastante (M) en una zona sin minería o industrias de Al. En las cercanías a la fábrica y en B1 el contenido de F fue alto en comparación a lo normal en plantas. Las diferencias en las concentraciones de F, Fe, Al y Pb por edad y sitio fueron significativas. En otras especies se ha reportado una correlación positiva de F con Al y fenoles totales, y negativa con Ca. En este trabajo se encontró, en B. crassifolia, una correlación positiva de F con Al, Ca y principalmente Fe, pero no con fenoles. Previamente no se había reportado la presencia de compuestos metálicos con F en esta especie, los cuales pueden ser liberados en infusiones medicinales. El lavado de hojas con agua mostró que una proporción alta de F y minerales estaba presente en la superficie foliar y/o los tricomas.

Summary

The concentration of fluoride (F) in relation to mineral elements (K, Ca, Mg, Mn, Fe, Cu, Zn, Ni, Co, Cr, Al, Pb and Cd) and total phenols was evaluated in leaves of Byrsonima crassifolia. Young, adult and fallen leaves were sampled in a garden located outside an aluminium factory in Bolivar State, Venezuela and in two sites in the savannah about 5 and 8km away from the Al-smelter in the direction of the wind. A positive linear regression of F with Mn, Fe, Pb, Al, Ca, Mg, Cr and Zn was found for young and adult leaves. Leaves were sampled in two more distant sites (B1, B2) and in a contrasting site (M) at a region without Al mining or factories. In the vicinity of the smelter and in B1 the content of F was high compared to the normal range in plants. The difference in concentrations of F, Fe, Al and Pb was highly significant by leaf age and site. In other species, a positive correlation of F to Al and total phenols, and a negative correlation to Ca has been reported, while in the present paper a positive correlation of F to Al, Ca and Fe, but not with total phenols was found in B. crassifolia. The presence of metal-fluorine compounds that can be released during medicinal infusions has not been previously reported in this species. Washing the leaves with water showed that a high proportion of F and minerals was present on the surface of the leaf and/or the trichomes.

Resumo

A concentração de fluoreto (F) em relação a elementos minerais (K, Ca, Mg, Mn, Fe, Cu, Zn, Ni, Co, Cr, Al, Pb y Cd) e fenóis totais foi avaliada em folhas de Byrsonima crassifolia. Se plantaram folhas jóvens, adultas e caídas em um jardim situado nas aforas de uma industria de alumínio, no Estado Bolívar, Venezuela e em dois lugares na savana a uma distância aproximada de 5 e 8 km da fábrica na direção do vento. Encontrou-se uma regressão lineal positiva de F com Mn, Fe, Pb, Al, Ca, Mg, Cr e Zn para folhas jovens e adultas. Se recolheram também folhas em outros dois lugares mais distantes (B1 e B2) e em um lugar contrastante (M) em uma zona sem mineria ou indústrias de Al. Nas cercanias da fábrica e em B1 o conteúdo de F foi alto em comparação ao normal em plantas. As diferenças nas concentrações de F, Fe, Al e Pb por idade e lugar foram significativas. Em outras espécies se tem reportado uma correlação positiva de F com Al e fenóis totais, e negativa com Ca. Neste trabalho encontrou-se, em B. crassifolia, uma correlação positiva de F com Al, Ca e principalmente Fe, mas não com fenóis. Previamente não se havía reportado a presença de compostos metálicos com F nesta espécie, os quais podem ser liberados em infusões medicinais. A lavagem de folhas com água mostrou que uma proporção alta de F e minerais estava presente na superfície foliar e/ou os tricomas.

Keywords / Aluminium / Fluoride / Iron / Nutrients / Pollution /

Received: 09/24/2003. Modified: 02/16/2004. Accepted: 02/27/2004.

Introduction

Fluorine is the most abundant halogen in the earth’s crust, but only six fluorinated natural products have been isolated, excluding fatty acid homologues (O’Hagan and Harper, 1999). The first organic fluorine compound identified, in 1943, was fluoracetate, which may cause death in grazing animals, because it is transferred to fluorocitrate and irreversibly bound to aconitase, thereby blocked the citrate cycle. Fluoracetate is found at low concentrations in a wide variety of plants, however it is accumulated at very high concentrations (421, 342 and 26mmol·kg^-1 on a dry weight basis, respectively) in seeds of Dichapetalum braunii from Tanzania, Gastrolobium bilobum from Australia and Palicourea marcgravii from South America, presumably for purposes of defense. Not only seeds can be toxic, as leaves of Oxylobium parviflorum from Australia contain up to 137mmol·kg^-1 fluoracetate (O’Hagan and Harper, 1999). According to Arnesen (1997) the maximum recommended limit for fluorine content in hay and pasture grass is 1.6mmol·kg^-1.

Inorganic fluorine is present in plants mainly as metal-fluorine compounds (Baroni Fornasiero, 2001). It is the most electronegative element and binds metals forming complexes (fluorides, hereunder F), which are adsorbed readily to the soil and plants.

Atmospheric pollution and repeated application of fertilizers and soil amendments, such as phosphatic fertilizers and phosphogypsum, increase the total F concentration in soils because they contain up to 3.5% of F as impurities (Stevens et al., 1997). A high correlation between F emission levels from Al-smelters in Norway and F accumulation in pine and spruce needles has been reported (Horntvedt, 1995).

Aluminium is the most prevalent cation with which F complexes in acid soils. Nagata et al. (1993) found evidence that those complexes are taken up and transported in tea plants until they reach the leaf where the complexes are dissociated. Fung et al. (1999) found that soil pH and extractable Al concentration affected the F contents in soils and plant tissues from tea plantations.

The present study was carried out on Byrsonima crassifolia, a medicinal plant distributed in several regions of Mexico, Central and South America, used since pre-Hispanic times. The bark and leaves are used to treat coughs, gastrointestinal disorders, skin infections and snake bites (Martínez-Vázquez et al., 1999). In Mexico it is the medicinal plant most frequently used against diarrhea (Leonti et al., 2001). Additionally, ethanol extracts of leaves showed some trypanocidal activity (Berger et al., 1998).

The dry bark and leaf of B. crassifolia is extracted in traditional medicine with ten volumes of water by infusion for 2h (Berger et al., 1998). It has been reported that very significant reductions in spontaneous locomotor activity and exploratory behavior are caused by a dosage of 1.25g dried plant/kg aqueous extracts of bark and leaves of B. crassifolia (Cifuentes et al., 2001).

It is important to know the F content of plants used in infusions. Fluorine is phytotoxic, causing damage in vegetation, wildlife and humans (Thomson et al., 1979; Weinstein and Davison, 2003). Fluorosis has been reported in inhabitants of Chinese provinces and Tibet where tea with a high F content is consumed (Fung et al., 1999, Cao et al., 2000). A small amount of F is beneficial in the prevention of dental caries and is used to treat osteoporosis, but high doses may cause damage.

A positive correlation of F to Al has been shown in tomato, oat, clover and Tibouchina pulchra (Arnesen, 1997; Stevens et al., 1997; Domingos et al., 2003), and a negative correlation to Ca in clover and ryegrass (Arnesen, 1997). The content of total phenols is positively correlated with F in Pinus nigra (Giertych et al., 1999).

The aim of this work was to study the concentration of F and the mineral composition of leaves of B. crassifolia, and to evaluate the differences in F concentration in relation to metals and total phenols, in leaves of different ages, in sites with pollution differences due to their proximity to or farness from an Al-smelter.

Material and methods

Plant material

Young (distal, light green, hairy and soft), adult (basal, green, leathery) and old fallen leaves (red-brownish, leathery) were sampled from five adult plants of Byrsonima crassifolia (L.) H.B.K. (Malpighiaceae) in each study site.

Study sites

Leaf samples were collected in a garden in the vicinity of an Al-smelter (8º16'N, 62º48'W) located near Ciudad Guayana, Bolivar State, Venezuela, and two sites in the savannah at a distance of 5km (8º15'N, 62º 50'W) and 8km (8º14'N, 62º51'W) west from the factory. The sampling was done in the prevailing wind direction (North-East) in May 2001.

Samples were also collected in two other sites in Bolívar State in May 2001. Site B1 (8º11'N, 62º52'W) was located 20km away from the Al-smelter in a savannah close to the intersection known as km 88 on the road from Ciudad Guayana to the iron mines of El Pao. Site B2 (8º10'N, 62º47'W) was located in Ciudad Guayana, at a university garden (Universidad Nacional Experimental de Guayana), 15km away from the Al-smelter but in opposite direction to the wind and smelter.

Comparisons were carried out with plants from site M, located at 10º8’N, 67º8'W in Miranda State, in a tree savannah without Al industries, where leaves were sampled in February, 2000.

Chemical analyses

Soluble fluoride (F) was determined in hot water extracts of dried ground leaves samples. A total of 100mg dry mass was infused with 5ml of distilled and deionized water using a heater block during approximately 1h until boiling was observed. The extracts were centrifuged at 2G for 10min after they were cooled to room temperature. The amounts of F in the solutions were determined by a F-specific ion electrode (Fluoride/fluoride combination electrode, model 96-09) and registered in an ion analyzer (Orion, model 720 A). The instrument was calibrated with standards of known concentrations of NaF in distilled and deionized water (Ionplus-Orion 0.1M). The supernatant was mixed 1:1 with TISAB-IV buffer to dissociate F-Complexes and stabilize pH. The buffer used contained 84ml concentrated HCl, 242g TRIZMA base and 230g sodium tartrate (Na₂C₄H₄O₆.2H₂O) diluted to 1l, according to the electrode instruction manual (Thermo Orion, 1999).

The leaves were oven-dried in paper sacks at 80ºC for 48h and ground. For the measurement of total K, Ca, Mg, Mn, Fe, Cu, Zn, Ni, Co, Al, Pb, Cr and Cd, 0.5g samples were digested in a nitric-perchloric acid mixture (Miller, 1998) and analyzed by flame atomic absorption spectrophotometry (SpectrAA 55B, Varian Techtron, Australia). The detection limits of the equipment were 2-800ppm Ca, 0.06-15ppm Fe, 0.15-20ppm Mg, 0.3-250ppm Al, 0.5-60ppm Mn, 0.01-2ppm Zn, 0.03-10ppm Cu, 0.1-20ppm Ni, 0.06-15ppm Cr, 0.1-30ppm Pb, 0.05-15ppm Co, and 15-800ppm K. Merck Standards were used for each element. With a peach leaves standard (reference 1547) from the National Institute of Standards and Technology, Gaithersburg, MD, a 96% Mg, 96% Ca, 99% Al, 97% Mn and 97% Cu recovery was obtained.

Total phenols were measured colorimetrically in ethanol extracts of ground dried leaves, using the Folin-Ciocalteu method (Amorim et al., 1977). Values were standardized against chlorogenic acid.

Washing experiment

Young leaves (n=18) were cut from the plants and washed with distilled and deionized water until the trichomes were not visible (1l per leaf). The same amount of water was used for the not hairy adult leaves (n=18). Opposite pairs of unwashed young and adult leaves were collected from the trees. The sampling was done in B1, B2 and in the Al-smelter garden on October 2002. Minerals and F were measured and the value obtained in the unwashed leaves was considered 100%. The difference between unwashed and washed was calculated as percent of the total in the leaf content and considered to be the F or mineral present on the leaf surface and trichomes (when they are present, as in young leaves). Thus, the distribution of F and minerals on the surface and in the leaf itself was indicated.

Statistics

Two-way ANOVA was done to compare the total concentration of mineral elements and phenols in B. crassifolia between three ages and three sites with Statistica 6.0 (Statsoft Inc.). Regression statistics were estimated with Sigma plot 2001 (7.0 S).

Results

Fluoride, metals and total phenols in different age leaves in relation to distance from an Al-smelter

In adult leaves the highest fluoride (F) concentration (7.0 ±0.1mmol·kg^-1) was obtained close to the factory (Figure 1). The concentration of Ca, Fe, Mg, Al, Mn (Figure 1), Zn, Cu, Ni, Cr, Pb and Co (Figure 2) was also high in those leaves in relation to the other samples. The decrease of F and metal contents observed in young and adult leaves in relation to the distance to the Al-smelter was not observed in fallen leaves for F, Ca, Al, Mn, Cu, Ni and Cr. The concentration of K in plants close to the factory was not higher than that of distant ones. Young leaves showed higher K contents than adult and fallen leaves. The leaves with the highest metal concentrations showed the lowest contents of phenols (Figure 2).

Table I shows the comparison by age, site and their interaction, of the differences in concentrations found for F, metals and phenols in the vicinity of the Al-smelter. Highly significant differences (p<0.001) were found for F, Ca, Fe, Al, Mn, Pb and phenols, but no differences were observed for Cu and Ni.

Positive linear regressions of F with Mn, Fe, Pb, Al, Ca, Mg, Cr and Zn were found for young and adult leaves (Figure 3, Table II). The power of the regression test with a= 0.0500 was 1, but for K, Cu, Ni, Co and total phenols it was below the desired power of 0.8 and the range of r was only 0.27-0.44. A negative slope for K and phenols was found. Cd was below the detection limit of 0.02ppm.

When all the study sites and foliar ages were analyzed together (data from Figures 1 and Table III) a positive correlation between Fe and F for young, adult and fallen leaves from the different sites were found, with r= 0.68, n= 121. With Ca, Mn, and Al the r was 0.49, 0.40 and 0.29 respectively. The power of the regression test with a= 0.05 was 1, 0.9970 and 0.9041, respectively, but for Pb, Mg, Zn, Ni and total phenols it was below the desired power of 0.8.

Values are means ±standard error, N=5. Sampling sites were B1: Savannah along theroad from Ciudad Guayana to iron mines at Pao (Bolívar State), located 20km from the Al- smelter, B2: Garden in a University campus in Ciudad Guayana (Bolívar State), and M: Savannah in Miranda State. Sampling was done in May 2001 in B1 and B2, and in February 2000 in M.

Fluoride, metals and phenols in leaves from sites with or without Al-industry

In old leaves from site B1 the concentration of F reached 13.7mmol·kg^-1. Lower values (Table III) were found at the site in the opposite wind direction of the smelter (B2) or at the site where no Al-industry exists (M). As expected, it was found that Ca and Mg concentration was higher than K concentration in adult and fallen leaves, but was not in young leaves from all sites. In M, but not in B1 nor B2, the concentration of Mg was higher than that of Ca. The micronutrient Fe was higher in B1, located at the road to the iron mines. The Al concentration reached high levels in B1 and B2. An increase of Mn concentration with age was observed. The concentration of phenols was >762mmol·kg^-1 and <1017mmol·kg^-1, excepting young leaves from site M and fallen leaves from B2, with lower values (529 and 541mmol·kg^-1, respectively).

Table IV shows the comparison by age, site and their interaction, of the differences in concentrations found for F, metals and phenols in the three sites studied. Highly significant differences (p<0.001) were found for F, Fe, Al and Pb, but no differences were observed for Cu.

Washing experiment

A large proportion of metals and F was located on the surface and/or trichomes in young leaves, and was therefore washed off with water (Figure 4) while only in the case of Cu no differences were observed in washed and unwashed leaves. Adult leaves showed lower proportions of Co, Al, Mn, Zn, Cr, Ca, K, Mg, Fe and Ni on the surface of the leaves than young leaves. However, 73% of F was washed away and was presumably present on the surface. In adult leaves Zn, Cr, Ni and Cu were not washed away, and for Ca, K, Mg and Fe the proportion present on the surface was <20%.

A higher proportion of Al was washed when compared to that of the macronutrients K, Ca and Mg, and the micronutrients Fe, Cu and Ni. For other micronutrients, such as Mn and Zn, the proportion washed in adult leaves was lower than that in young leaves.

Discussion

High fluoride content in samples collected at an industrial site

Background concentration of F in plants is usually <0.5mmol·kg^-1 according to Arnesen (1997). Normal F content of leaves generally ranges from 0.1 to 1mmol·kg^-1 (Baroni Fornasiero, 2001). Young, adult and fallen leaves of B. crassifolia sampled at the garden of the factory and 5km away were more concentrated in F than the normal range, but at 8km from the factory only fallen leaves were contaminated with F (Figure 1).

Murray (1985) also showed high concentrations of F (32mmol·kg^-1) in leaves of Avicennia marina (grey mangroves), generally without visible injury, in a study site located approximately 100m from the Australian Fluoride Chemicals plant.

In an industrial complex in Brazil, which includes fertilizer industries, steel works, refineries, chemical and petrochemical plants, a foliar F concentration of 36mmol·kg^-1 was found in Lolium multiflorum (Klumpp et al., 1994), a species used as a bioindicator in Europe and North America because of its accumulation of toxic elements (F, S, metals). Domingos et al. (2003) suggested that Al-F complexes were taken from the soil in the Al-accumulator tree species Tibouchina pulchra in this industrial complex in Brazil, where plants are exposed to fluoride-contaminated air.

Hydrogen fluoride is usually associated with the production of aluminium, steel, brick, tile, phosphate fertilizer, recycling of uranium fuel and coal combustion. Metallic aluminium is produced by the electrolytic reduction of alumina, which is produced from bauxite and in this process alumina dissolved in cryolite is emitted together with gas containing F. The vegetation in the vicinity of aluminium factories accumulated high F contents in Greece (Malea, 1995).

No visible symptoms of damage were observed in B. crassifolia with high F concentrations. Therefore, this species is not suitable as a F bioindicator. Weinstein and Davison (2003) described visible symptoms of injury in sensitive species exposed to F-emitting sources, specially Al-smelters and phosphate fertilizer plants, and listed 24 native species from South America suitable as bioindicators for the presence of airborne fluorides. The use of bioindicators is described as a rapid and inexpensive technique that provides an estimate of the degree of injury at the time of survey.

Fluoride in relation to age

The content of F in young leaves was lower than in adult or fallen leaves (Figure 1 and Table III), as was found in brick tea (Fung et al., 1999). Leaf age strongly affected the F content in leaves in Salix caprea and Betula pendula (Vike and Håbjørg, 1995).

In Pinus sylvestris (Scots pine) an increase in concentration from needle base to tip was detected for Al, Fe and F (Giertych et al., 1997). This distal accumulation of toxic elements led to necroses. Giertych et al. (1999) also showed an increase of content of F with age of needles (5 classes) in Pinus nigra.

High Al contents in leaves but without hyperaccumulation

The concentration of Al in plants from sites B1 and B2 was higher than in those from M (Table III). The concentration of Al was high in adult leaves sampled at the factory garden (36mmol·kg^-1; Figure 1), and in adult and fallen leaves collected in the garden of the University of Guayana (30 and 25mmol·kg^-1, respectively; Table III). Higher values have been observed in Al accumulators such as Melastoma malabathicum (Melastomataceae) in old leaves sampled from greenhouse grown plants (534mmol·kg^-1; Watanabe et al., 1998), Richeria grandis (Euphorbiaceae) from a cloud forest in Venezuela (556mmol·kg^-1; Cuenca et al., 1991), Craterispermum laurinum (Rubiaceae) from a savannah in Liberia (1352mmol·kg^-1; Jansen et al., 2003), Maschalocorymbus corymbosus sp. (Rubiaceae) from a tropical rain forest in Indonesia (1368mmol·kg^-1; Masunaga et al., 1998) and in mature leaves of Faramea marginata (Rubiaceae) from a swamp forest in Brazil (767mmol·kg^-1; Britez et al., 2002). According to Jansen et al. (2002) the largest recorded content of aluminium in any plant is 2677mmol·kg^-1, in Symplocos spicata (Symplocaceae).

The concentration of Al in Byrsonima crassa, B. coccolobifolia and B. verbascifolia found by Haridasan (1982) in the tree savannahs (cerrado region) of central Brazil was between 8 and 13mmol·kg^-1, where other species such as Vochysia thyrsoidea reached 523mmol·kg^-1.

High Fe contents in leaves from Bolivar samples

Fe in adult leaves from the factory garden (Figure 1) was very high (125 ±7mmol·kg^-1) and lower values (<76 ±6mmol·kg^-1) were found in the other study sites (Table III). Fe was not measured in the genus Byrsonima from the savannahs by Haridasan (1982), but in other species from the cerrado region, such as Miconia ferruginata it was reported at 4mmol·kg^-1. Jansen et al. (2003) reported 12mmol·kg-1 in Melastoma malabathricum and 13mmol·kg^-1 in Coccocypselum canescens. In a crop species, such as Spinacia oleracea, considered to contain high Fe levels, 9mmol·kg^-1 has been reported (Bhattacharjee et al., 1998). However, in Eichhornia crassipes 179mmol·kg^-1 Fe have been found (Larcher, 1995) and Clark and Baligar (2000) indicated that plants normally accumulate relatively high Fe before disorders such as bronzing or brown speckling appear, increasing from 5.4 to 12.5mmol·kg^-1 in rice and from 19.7 to 28.6mmol·kg^-1 in Rumex.

Total cation and phenols contents in trees from different sites in relation to micronutrients

The sum of the cation (K, Ca, Mg, Fe, Mn and Al) contents was >416mmol·kg^-1 in fallen leaves from the different sites (B1, B2, M), but <295mmol·kg^-1 in young leaves (Table III). These values in B. crassifolia from savannahs with oligotrophic conditions resulted lower than the total cation content, between 500 and 700mmol·kg^-1, for most of the species studied by Köhl et al. (1996) in the Canary Islands.

In Bolívar State, Al and Fe mining, and industrial related activities, take place. This can explain the high levels in foliar Al and Fe concentration reached, in comparison to those in M, a site without these activities but where Ni mining takes place. Accordingly, leaves showed higher Ni contents in M (Table III).

The Fe/Al molar ratio in the proximity of the Al factory (calculated from Figure 1) was between 1.96 and 3.51, and F concentration was >3.16mmol·kg^-1, except in young leaves (at 0, 5 and 8km) or in adult leaves from trees distant 8km from the factory. In B1 the Fe/Al ratio (Table III) increased with age, to a maximum in fallen leaves (Fe/Al= 3.7 and F= 13.68mmol·kg^-1). In B2 and M the Fe/Al ratio was <1.23 and F <1.29mmol·kg^-1.

The concentration of phenols was high in all the samples (Figure 2 and Table III) as compared with the range reported by Coley (1983) in tropical plants (48-1098mmol·kg^-1).

The content of total phenols was positively correlated with F in Pinus nigra (black pine; Giertych et al., 1999). However, in the present study such a relation was not observed between phenols and F in B. crassifolia. On the other hand, Baroni Fornasiero (2001) showed by electron microscopy many dark stained, vacuolar, electron dense globular inclusions in Hypericum perforatum (Clusiaceae), interpretable as condensed tannins (phenols), which were scarcely visible after F-treatment, but abundant in the leaf mesophyll cells, especially in the control plant leaves. Baroni Fornasiero (2001, 2003) reported increases in anthocyanin and hypothesized that in H. perforatum, where the proanthocyanidins or condensed tannins are abundant, the red-brown pigmentation of the F-affected areas is related to the increased amounts of anthocyanins during the F-induced stress.

Washing experiment

A high proportion of F and metals was found to be on the surface and/or trichomes of the leaves (Figure 4) and could be washed off by rain. Isermann (1977) showed Pb deposited by car exhausts on Lolium perenne was easily washed (for 15min with distilled water) reducing the lead content by 27%, while washing Betula pubescens leaves with citric acid, pH 2.5, reduced the F content by 15% (Vike and Håbjørg, 1995).

The young leaves were washed until trichomes were not present anymore. Several examples in the literature showed the role of trichomes in metal tolerance, such as Pb in Nicotiana tabacum (Martell, 1974), or Mn in Helianthus annuus (Blamey et al., 1986). Trichome glands were found in the petioles of Nymphaea sp., where polyphenols play a role in metal detoxification (Lavid et al., 2001 a, b).

Conclusions

B. crassifolia leaves, sampled in industrial areas where Al is processed, showed high levels of F associated to metals. F was not only associated to Al, as reported previously, but mainly to Fe. These mineral elements may be present in the medicinal infusions, even when leaves are previously washed with water.

REFERENCES

1. Amorim HV, Dougall DK, Sharp WR (1977) The effect of carbohydrate and nitrogen concentration on phenol synthesis in Paul’s Scarlet Rose cells grown in tissue culture. Physiol. Plant. 39: 91-95.        [ Links ]

2. Arnesen AKM (1997) Availability of fluoride to plants grown in contaminated soils. Plant Soil 191: 13-25.        [ Links ]

3. Baroni Fornasiero R (2001) Phytotoxic effects of fluorides. Plant Sci. 161: 979-985.        [ Links ]

4. Baroni Fornasiero B (2003) Fluorides effects on Hypericum perforatum plants: first field observations. Plant Sci. 165: 507-513.        [ Links ]

5. Berger I, Barrientos AC, Cáceres A, Hernández M, Rastrelli L, Passreiter CM, Kubelka W (1998) Plants used in Guatemala for the treatment of protozoal infections. II. Activity of extracts and fractions of five Guatemalan plants against Trypanosoma cruzi. J. Ethnopharmacol. 62: 107-115.        [ Links ]

6. Bhattacharjee S, Dasgupta P, Paul AR, Ghosal S, Pahdi KK, Pandey LP (1998) Mineral element composition of spinach. J. Sci. Food Agricult. 77: 456-458.        [ Links ]

7. Blamey FPC, Joyce DC, Edwards DG, Asher CJ (1986) Role of trichomes in sunflower tolerance to manganese toxicity. Plant Soil 91: 171-180.        [ Links ]

8. Britez RM, Watanabe T, Jansen S, Reissmann CB, Osaki M (2002) The relationship between aluminium and silicon accumulation in leaves of Faramea marginata (Rubiaceae). New Phytol. 156: 437-444.        [ Links ]

9. Cao J, Zhao Y, Liu J, Xirao R, Danzeng S (2000) Environmental Fluoride content in Tibet. Environ. Res. 83: 333-337.         [ Links ]

10. Cifuentes C, Gómez-Serranillos MP, Iglesias I, Villar del Fresno AM (2001) Neuropharmacological profile of ethnomedicinal plants of Guatemala. J. Ethnopharmacol. 76: 223-228.        [ Links ]

11. Clark RB, Baligar VC (2000) Acidic and alkaline soil constraints on plant mineral nutrition. In Wilkinson RE (Ed.) Plant-Environment Interactions. Marcel Dekker. New York, USA. pp. 133-177.        [ Links ]

12. Coley PD (1983) Herbivory and defensive characteristics of tree species in lowland tropical forest. Ecol. Monographs 53: 209-233.        [ Links ]

13. Cuenca G, Herrera R, Mérida T (1991) Distribution of aluminium in accumulator plants by x-ray microanalysis in Richeria grandis Vahl leaves from a cloud forest in Venezuela. Plant Cell Env. 14: 437-441.        [ Links ]

14. Domingos M, Klumpp A, Rinaldi MCS, Modesto IF, Klumpp G, Delitti WBC (2003) Combined effects of air and soil pollution by fluoride emissions on Tibouchina pulchra Cogn., at Cubatão, SE Brazil, and their relations with aluminium. Plant Soil 249: 297-308.        [ Links ]

15. Fung KF, Zhang ZQ, Wong JWC, Wong M (1999) Fluoride contents in tea and soil from tea plantations and the release of fluoride into tea liquor during infusion. Environ. Pollut. 104: 197-205.        [ Links ]

16. Giertych MJ, De Temmerman LO, Rachwal L (1997) Distribution of elements along the length of Scots pine needles in a heavily polluted and a control environment. Tree Physiol. 17: 697-703.        [ Links ]

17. Giertych MJ, Karolewski P, De Temmerman LO (1999) Foliage age and pollution alter content of phenolic compounds and chemical elements in Pinus nigra needles. Water Air Soil Pollut. 110: 363-377.        [ Links ]

18. Haridasan M (1982) Aluminium accumulation by some cerrado native species of central Brazil. Plant Soil 65: 265-273.        [ Links ]

19. Horntvedt R (1995) Fluoride uptake in conifers related to emissions from aluminium smelters in Norway. The Science of the Total Environ. 163: 35-37.        [ Links ]

20. Isermann K (1977) A method to reduce contamination and uptake of lead by plants from car exhaust gases. Environ. Pollut. 12: 199-203.        [ Links ]

21. Jansen S, Broadley MR, Robbrecht E, Smets E (2002) Aluminum hyperaccumulation in Angiosperms: a review of its phylogenetic significance. Bot. Rev. 68: 235-269.        [ Links ]

22. Jansen S, Watanabe T, Dessein S, Smets E, Robbrecht E (2003) A comparative study of metal levels in leaves of some Al-accumulating Rubiaceae. Ann. Bot. 91: 657-663.        [ Links ]

23. Klumpp A, Klumpp G, Domingos M (1994) Plants as bioindicators of air pollution at the Serra do Mar near the industrial complex of Cubatão, Brazil. Environ. Pollut. 85: 109-116.        [ Links ]

24. Köhl K, Lösch R, González-Rodríguez AM, Jiménez MS, Morales D (1996) Foliar cation contents of laurel forest trees on the Canary Islands. Flora 191: 303-311.        [ Links ]

25. Larcher W (1995) Physiological Plant Ecology. Springer. Berlin, Germany. 506 pp.        [ Links ]

26. Lavid N, Barkay Z, Tel-Or E (2001a) Accumulation of heavy metals in epidermal glands of the waterlily (Nymphaeaceae). Planta 212: 313-322.        [ Links ]

27. Lavid N, Schwartz A, Yarden O, Tel-Or E (2001b) The involvement of polyphenols and peroxidase activities in heavy-metal accumulation by epidermal glands of the waterlily (Nymphaeaceae). Planta 212: 323-331.        [ Links ]

28. Leonti M, Vibrans H, Sticher O, Heinrich M (2001) Ethnopharmacology of the Popoluca, Mexico: An evaluation. J. Pharm. Pharmacol. 53: 1653-1669.        [ Links ]

29. Malea P (1995) Fluoride content in three seagrasses of the Antikyra Gulf (Greece) in the vicinity of an aluminum factory. The Sci. of the Total Environ. 166: 11-17.        [ Links ]

30. Martell EA (1974) Radioactivity of tobacco trichomes and insoluble cigarette smoke particles. Nature 249: 215-217.        [ Links ]

31. Martínez-Vázquez M, González-Esquinca AR, Cazares-Luna L, Moreno-Gutiérrez MN, García-Argáez AN (1999) Antimicrobial activity of Byrsonima crassifolia (L.) H.B.K. J. Ethnopharmacol. 66: 79-82.        [ Links ]

32. Masunaga T, Kubota D, Hotta M, Wakatsuki T (1998) Mineral composition of leaves and bark in aluminium accumulators in a tropical rain forest in Indonesia. Soil Sci. Plant Nutr. 44: 347-358.        [ Links ]

33. Miller RO (1998) Nitric-perchloric acid wet digestion in an open vessel. In Kalra YP (Ed.) Handbook of Reference Methods for Plant Analysis. CRC Press. Boca Raton USA. pp. 57-61.        [ Links ]

34. Murray F (1985) Cycling of fluoride in a mangrove community near a fluoride emission source. J. Appl. Ecol. 22: 277-285.        [ Links ]

35. Nagata T, Hayatsu M, Kosuge N (1993) Aluminium kinetics in the tea plant using Al-27 and F-19 NMR. Phytochemistry 32: 771-775.        [ Links ]

36. O’Hagan D, Harper DB (1999) Fluorine-containing natural products. J. Fluorine Chem. 100: 127-133.        [ Links ]

37. Stevens DP, McLaughlin MJ, Alston AM (1997) Phytotoxicity of aluminium-fluoride complexes and their uptake from solution culture by Avena sativa and Lycopersicon esculentum. Plant Soil 192: 81-93.        [ Links ]

38. Thermo Orion (1999) Fluoride/fluoride combination electrode. Instruction Manual. Orion Research Inc. Beverly, MA, USA. 42 pp.        [ Links ]

39. Thomson LK, Sidhu SS, Roberts BA (1979) Fluoride accumulation in soil and vegetation in the vicinity of a phosphorus plant. Environ. Pollut. 18: 221-234.        [ Links ]

40. Vike E, Håbjørg A (1995) Variation in fluoride content and leaf injury on plants associated with three aluminium smelters in Norway. The Science of the Total Environ. 163: 25-34.        [ Links ]

41. Watanabe T, Osaki M, Yoshihara T, Tadano T (1998) Distribution and chemical speciation of aluminum in the Al accumulator plant, Melastoma malabathricum L. Plant Soil 201: 165-173.        [ Links ]

42. Weinstein LH, Davison AW (2003) Native plant species suitable as bioindicators and biomonitors for airborne fluoride. Environ. Pollut. 125: 3-11.        [ Links ]