AOBPreview originally published online on July 2, 2004
Annals of Botany 2004 94(2):259-268; doi:10.1093/aob/mch136
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Annals of Botany 94/2, © Annals of Botany Company 2004; all rights reserved
Effects of Genetic Variability and Habitat of Qualea parviflora (Vochysiaceae) on Herbivory by Free-feeding and Gall-forming Insects
1 Pós-Graduação em Ciências Genômicas e Biotecnologia, Universidade Católica de Brasília, SGAN 916, Mod. B, Brasília, DF, 70790-160, Brazil and 2 Ecologia Evolutiva de Herbívoros Tropicais, DBG/UFMG, CP 486, Belo Horizonte, MG 30161-970, Brazil
* For correspondence. E-mail silmaryalvim{at}uol.com.br
Received: 28 November 2003 Returned for revision: 28 January 2004 Accepted: 8 April 2004 Published electronically: 2 July 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
• Background and Aims Differences in the chemical and physical traits of plants caused by both genetic and habitat characteristics may influence attack by herbivores. Leaves of Qualea parviflora (Vochysiaceae), a common tree of different habitats in the Brazilian Neotropical savannas (cerrado), are susceptible to severe attack by herbivorous free-living and gall-forming insects. Attack by free-living and gall-forming insects within and between populations of Q. parviflora were examined and it was determined to what extent genetic variability (detected by RAPD markers), phenotypic characteristics of the plants and habit traits influence the number of free-living and gall-forming insect species and individuals attacking the plants, and the intensity of attack.
• Methods On four occasions in 2000, leaves were sampled from ten individual trees in each of three types of vegetation in the cerrado: campo sujo, cerrado sensu stricto and cerradão at the Ecological Station of Pirapitinga (ESP), in Três Marias, north-western Minas Gerais, Brazil. Genetic variability was detected by RAPD markers, and concentrations of nutrients, phenols and tannins, sclerophylly and pre-dawn water potential of leaves were measured. Water and nutrient contents in the soil below each tree characterized the habitat. The free-living and gall-forming herbivorous insects were determined.
• Key Results Of 69 RAPD markers analysed, 41 were polymorphic and were used for analyses of genetic variability of Q. parviflora. Most of the variability occurred within habitats, accounting for 97·65 % of the genetic variability. Plants in the cerrado sensu stricto and campo sujo were the most similar. There were no significant associations between genetic similarity and the chemical and physical traits of Q. parviflora, or with habitat, nor was there significant correlation between phenotypic and habitat traits. Increasing concentrations of tannins and sulphur, C : N ratio and sclerophylly correlated with increasing percentage of leaf area damaged by herbivores. Decreased sclerophylly, concentration of tannins and C : N ratio, and increased concentration of nutrients in leaves correlated with increased severity of attack by gall-forming insects.
• Conclusions Nutrient concentration in the soil had more influence, indirectly, on free-feeding insects than did composition of Q. parviflora leaves. However, gall-forming insects are affected more by leaf quality, attacking fewer sclerophyllous leaves, with larger nutrient but smaller tannin concentrations.
Key words: Cerrado, genetic variability, gall attack, herbivory, insect galls, plant quality, Qualea parviflora, RAPD, Vochysiaceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Variation in the resistance of individual plants to attack by herbivorous insects may be caused by differences in genotype and in the plant's environment (Mutikainen et al., 2000
). Intra-population genotypic differences in resistance have been observed in several species (e.g. Anderson et al., 1989; Fritz, 1990; Karban, 1992; Zangerl and Berenbaum, 1997). These differences in the host plant's resistance can exert selective pressures on the ability of insects to determine variations in the plant's characteristics favourable to them (Anderson et al., 1989). Conversely, differences in herbivory can affect plant fitness, often having negative effects by decreasing growth rate, reproductive success and competitive ability, all of which can affect the genetic structure of populations (Coley, 1983; Coley and Barone, 1996).
The susceptibility of plants to attack by herbivores may be influenced by the plant's quality, which in turn is linked to factors such as soil nutrients, light intensity and past herbivory (Bazzaz et al., 1987). Intraspecific variation in defence against herbivory may be influenced by spatial and temporal variation in resource availability, and to the physiological condition of the plants, such as size and age (Chapin et al., 1987). Some habitat features, such as composition and distribution of host plant species and composition and abundance of natural enemies, may also influence the activity of herbivores by modifying their patterns of search and colonization (Bazzaz et al., 1987).
Although the acceptance of a site for oviposition (as in gall-forming insects) or the amount of food ingested (as in free-feeding herbivorous insects) also depends on the physiological status of the individual insect (such as food supply and deprivation, egg load, age) and its experience (Schoonhoven et al., 1998), the effect of the host-plant on insect performance (growth, survival and fecundity) is the major determinant of choice (Hartnett and Abrahamson, 1979). For example, the adults of cabbage white butterfly, Pieris rapae, selectively laid eggs on Brassica oleracea (Brassicaceae) plants with higher nitrogen and phosphorous content, on which their caterpillars grew faster and attained a heavier weight (Myers, 1985). Stein and Price (1995) also observed that females of two gall-forming tenthredinid sawflies, Euura sp. and Pontania sp., avoid shoots of very young ramets of Salix lasiolepis Benth. (Salicaceae) on which survival of their offspring is decreased.
During the search for a host plant, herbivorous insects behave in particular ways, such as repetitive contacts of legs, antennae, mouthparts or ovipositor with the plant surface, or even take a test bite; these serve to evaluate the physical and chemical characteristics of the leaf and determine acceptance or refusal (Schoonhoven et al., 1998). Physical features of plant organs or tissues which affect feeding by insects include the presence of trichomes and wax crystals on the surface, and the thickness and toughness (sclerophylly and high silica content) of the tissue (Ernest, 1989). For example, Woodman and Fernandes (1991) showed that Verbascum thapsus (Scrophulariaceae) leaves varied in density of hairs and demonstrated that generalist herbivores selected the least pubescent leaves as opposed to those which are very pubescent. In adidition, cicadellids (Homoptera) and tingids (Hemiptera) chose to feed on leaves of the tropical tree Tabebuia ochracea (Bignoniaceae) with small concentrations of calcium oxalate crystals (Ribeiro et al., 1994). Coley (1983) found that leaf toughness was the most important factor determining leaf palatability for herbivores on persistent and pioneer trees of the tropical rain forest in Panama.
Nutrient concentration and type, and concentration of compounds such as tannins and other phenolics (generally regarded as secondary defence metabolites) in leaves are examples of chemical characteristics that influence insect behaviour (Whittaker and Feeny, 1971; Schoonhoven et al., 1998). Marquis et al. (2001) found that phenolics have an important role as defences against both pathogens and insects in many plant species of the Brazilian cerrado. Cornelissen and Fernandes (2001) also suggested that the area of leaf damaged by herbivores in Bauhinia brevipes (Leguminosae) can be influenced by, for example, water content, sugars, toughness, and nitrogen content of the leaves.
Environmental factors affecting the plant and its response to herbivory are extremely variable, and poorly understood. The effects of these factors were examined on Qualea parviflora (Vochysiaceae), which is a deciduous cerrado tree species that accumulates aluminium (to leaf Al concentrations above 10 000 mg kg–1; Haridasan, 1982), and is distributed in different ecological regions of Brazil. There the cerrado (Neotropical savanna) is the second most important biome, covering about 20 % of the country (Felfili and Silva, 1993). Three floristic types (called physiognomies) are commonly found: (1) campo sujo, a savanna-like vegetation, with scattered trees and shrubs and canopy cover of <2 %; (2) cerrado sensu stricto, a savanna-like vegetation with a higher density of trees and shrubs (canopy cover of about 20 %): (3) cerradão, a xeromorphic forest vegetation with a fairly continuous canopy tree, which ranges from 15 to 85 % cover (Eiten, 1972; Felippe and Dale, 1990). Qualea parviflora commonly occurs in these three types of vegetation, where free-living herbivorous and gall-forming insects severely attack the leaves.
The following were studied: (a) variation in the number of species of herbivorous free-living and gall-forming insects; (b) the intensity with which the insects attack Q. parviflora leaves; and (c) the extent to which genetic variability [detected by random amplified polymorphic DNA (RAPD) markers], and plant end environmental characteristics influence the insects. The following questions were addressed: How is the genetic variability of Q. parviflora distributed within and among populations in the three types of cerrado vegetation? How are genetic variability, phenotype and habitat of the host plant related? Which factors best explain variations in the number of individuals of herbivorous free-living and gall-forming insect species and the intensity with which they attack host plants?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Study site
The study was carried out at the Ecological Station of Pirapitinga (ESP), 1000 ha of cerrado biome in south-east Brazil (18°20'S to 18°23'S and 45°17'W to 45°20'W), where three types of vegetation occur: campo sujo, cerrado sensu stricto and cerradão. The distribution of Q. parviflora Mart. (Vochysiaceae) was continuous over the area but plant density varied in each vegetation type: 0·12 trees m–2 in cerradão, 0·34 trees m–2 in campo sujo and 0·39 trees m–2 in cerrado sensu stricto. The distances between the populations in cerrado sensu stricto and campo sujo and cerradão were approx. 142 and 202 m, respectively, and cerradão and campo sujo were about 172 m apart.
Genetic variability in Q. parviflora
For genetic variability analysis, undamaged, young leaves from each of 30 plants were collected in October 2000. Leaves were washed in distilled water and stored in a freezer at –70 °C. DNA was extracted from about 100–150 mg of leaf tissues using the standard method of Doyle and Doyle (1987).
Initially, 24 primers of Operon Tecnologies Inc. (CA, USA) were screened for polymorphism and amplification quality using eight randomly chosen individuals from the three populations. RAPD amplifications were performed in a 13 µL volume containing 1x reaction buffer (10 mM Tris–HCl, pH 8·3, 50 mM KCl, 1·5 mM MgCl2), 1 unit Taq DNA polymerase (Invitrogen, MD, USA), 200 µM of each dNTP, 30 ng of primer, 3·25 µg bovine serum albumin and 9·0 ng of template DNA. Amplifications were performed using a GeneAmp PCR System 9700 (Applied Biosystems, CA, USA) under the following conditions: 94 °C for 5 min (1 cycle); 94 °C for 1 min, 35 °C for 1 min and 72 °C for 2 min (40 cycles); and 72 °C for 7 min (one cycle). Polymorphism was detected using 1·5 % agarose gels stained with ethidium bromide (1·0 mg mL–1) and sized by comparison to a 1 kb Plus DNA ladder standard (Invitrogen).
Primers that amplified polymorphic and clearly interpretable fragments were selected to amplify the individuals from the three populations (Table 1). RAPD amplifications followed the protocols described, and the products were electrophoresed for 4 h in 1·5 % agarose gels, stained with ethidium bromide (1·0 mg mL–1) and the size of the products determined by comparison to a 1 kb Plus DNA ladder standard (Invitrogen). Gels were photographed using Electrophoresis Documentation and Analysis System 120 (Kodak Digital ScienceTM) and analysed visually, on a computer screen, for presence or absence of bands. Only bands of high intensity were included in the analyses. Poorly amplified or very sporadic bands were not considered.
|
Phenotypic traits
To study phenotypic traits of Q. parviflora, 53 fully expanded and undamaged leaves were collected from each of the 30 plants in January, April, July and November in 2000. To determine within- and between-population variation in phenotypic traits, the four samples taken at different dates were pooled.
The mean concentration of macro- and micronutrients, total phenols (percentage in dry matter) and tannins (percentage of dry matter) were measured for each of the 30 plants. A different number of leaves per individual plant was collected for nutrient analysis (number of leaves per plant n = 20), total phenols (n = 10), sclerophylly (n = 20) and tannins (n = 3) because of differences in methods. The macro- and micronutrients were analysed on 400 mg of dry plant tissue after wet digestion with HNO3 and HClO4. Nitrogen was analysed by flow injection analysis (Carlson et al., 1990). Phosphorous was determined colorimetrically (Silva, 1999). Carbon was determined indirectly by the Walkely and Black method (Walkely and Black, 1934). An air–acetylene flame spectrophotometer was used for sodium and potassium determination (Johnson and Ulrich, 1959). Silica was analysed colorimetrically (Elliott and Snyder, 1991). Calcium, sulphur, magnesium, aluminium, iron, copper, zinc and manganese were measured by atomic absorption spectrophotometry (Meyer and Keliher, 1992).
Condensed and hydrolysable tannins were measured (three-fold replication per plant) in fresh leaves by the radial diffusion method with tannic acid as standard (Hagerman, 1987). For total phenols, the leaves were air-dried, triturated and analysed using the method of Swain (1979). Sclerophylly index (g cm–2) was calculated in 20 fully expanded and non-damaged leaves for each plant (n = 30 trees). Leaves were drawn and scanned to measure the leaf area (Scion-Image software, 1998), and then were oven-dried at 45 °C for at least 72 h, after which the mass of the dry matter component was determined. In addition, pre-dawn (0300–0600 h) leaf water potential (MPa) was measured with a Scholander pressure chamber (PMS-600) on one shoot per plant in each sampling date (Scholander et al., 1964, 1965).
Habitat features
A sample of soil was removed to 20 cm in depth under each individual tree in each vegetation type (n = 30 trees). The soil samples were air-dried and the following analysed: N-ammonium (%), N-nitrate (%), P (mg dm–3), Ca (mg dm–3), K (mg dm–3), Mg (mg dm–3), organic matter (dag kg–1), Si (mg kg–1), Na (mg dm–3), Fe (mg dm–3), Cu (mg dm–3), Zn (mg dm–3), Mn (mg dm–3), Al (cmolc dm–3), exchangeable acidity [H + Al (mg dm–3)], total and effective cation exchange capacities, aluminium saturation (m%), pH in water, and soil water content (%).
Soil water content was determined by difference between fresh and dry weight (dried for 72 h, or to constant weight, at 45 °C). Phosphorous was determined colorimetrically (Silva, 1999). Organic matter was determined indirectly by the Walkely and Black method (Walkely and Black, 1934). Inorganic nitrogen was analysed by the adapted methods of Mullin and Riley (1955) and Kempers and Zweers (1986). An air–acetylene flame was used for potassium (Johnson and Ulrich, 1959). Silica was analysed colorimetrically (Hallmark et al., 1982). Calcium, iron, copper, zinc, manganese and aluminium concentrations were measured by atomic absorption spectrophotometry (Meyer and Keliher, 1992).
Sampling of adult gall-forming and free-living insects and leaf damage
Adult herbivorous insects associated with Q. parviflora leaves were sampled in the morning (between 0800 and 1200 h) and in the afternoon (between 1400 and 1800 h), for 10 min per plant on each collection date, in each population. All the insects observed on the leaves were collected, and sent to specialists for identification.
For each tree, ten shoots were randomly collected around the circumference of each plant (n = 300 shoots). For each shoot and plant the total number of leaves, number of leaves attacked by insects, and gall abundance (new and hatched galls) were recorded, and means per plant were estimated. Free-living insect damage to the foliage was quantified in two mature leaves per shoot (20 leaves per plant). Leaves were drawn and digitized by a flat-bed scanner. The total leaf area and percentage of leaf area damaged were calculated using the Scion-Image software (1998). Only herbivore damage that resulted in lamina loss (leaf removal, chewing or mining) was included (see Moles and Westoby, 2000).
Data analysis
To verify how genetic variability of Q. parviflora is distributed within and between populations the raw data matrix (presence and absence of bands) was used to generate a similarity matrix based on the Jaccard coefficient (Sneath and Sokal, 1973), where J = 0 is considered total dissimilarity and J = 1 is total similarity. An UPGMA cluster analysis was performed based on the similarity matrix using the software NTSYS version 2·02k (NTSYS, 1997). Additionally, an analysis of molecular variance (AMOVA; Excoffier et al., 1992) was performed using the Arlequin Ver 2·0 software (Schneider et al., 2000).
To verify the relationship between genetic variability, phenotype and habitat characteristics, a Mantel test (Mantel, 1967) was made using NTSYS v. 2·02k (NTSYS, 1997). As phenotype and habitat variables had different units, before analysis all variables were standardized by the following formula: [(x – Mx)/SDx], where M = mean and SD = standard deviation (NTSYS, 1997). A similarity matrix was obtained using the Jaccard coefficient for qualitative data (Sneath and Sokal, 1973).
Principal component analysis (PCA) was done using a correlation matrix of raw data to establish the relationship among genetic variability, and phenotypic and habitat traits of Q. parviflora, using Systat v. 8·0. Initial PCA showed that some variables did not have factor loadings greater than 0·50 on retained components. These variables were then taken out from the analysis and the PCA was repeated (see Marquis et al., 2001).
Multiple linear correlation and step-wise multiple linear regression were used to test which factors are significant in influencing variation in attacks by free-living herbivores and gall-formation on the host-plant (Zar, 1996). Herbivorous attack was estimated by the number of attacked leaves and percentage of leaf area damaged, while gall-forming insect attacks were estimated from the number of leaves with galls and number of galls per individual plant. Afterwards, simple and multiple linear regressions were used to verify the relationship between number of species and attack by free-living herbivores and gall-forming insects and the new axis obtained from PCA (see Collevatti and Schoereder, 1995; Marquis et al., 2001).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Genetic variability in Q. parviflora
Sixty-nine polymorphic RAPD markers were detected using five selected primers, and 41 polymorphic bands (6·8 bands per primer) with better resolution were used in subsequent analyses (Table 1). From these, 80·5 % were found in plants of all three sampling sites, 14·6 % were present at two sites, and 4·9 % were only found in one type of vegetation.
The results showed large genetic similarity among plants from different populations (Fig. 1). The highest similarity (J = 0·82) occurred between two individuals from different types of vegetation, the cerrado sensu stricto and campo sujo (individuals 17 and 21; Fig. 1). Seven clusters could be identified: I—with plants from the three populations; II—with plants from cerrado sensu stricto and cerradão; III—with plants from cerradão; IV—with a plant from cerradão and others from the cerrado sensu stricto; V—with a plant from cerradão and others from the cerrado sensu stricto; VI—with a plant from cerradão and others from campo sujo; and VII—with plants from the campo sujo. Individuals 2, 13, 25 and 26 had similarity coefficients lower than 0·50 with other clusters (Fig. 1).
|
The analysis of molecular variance indicated that there was no significant difference between the ten trees selected in each of the three populations in different habitats (Table 2). Most of the genetic variation (97·65 %) of Q. parviflora is distributed among individuals of the same population (Table 2).
|
Relationship among genetic variability, phenotypic and habitat traits
A PCA based on presence and absence of RAPD markers on plants from the three types of vegetation showed that 56·6 % of the variation could be explained by four axes, and only 18 RAPD markers had a correlation above 0·50 with one of the new PCA axes. Furthermore, there was no significant correlation between the genetic similarity of individual Q. parviflora trees and the chemical and physical characteristics of the plants (Mantel test: r = –0·026; P = 0·395) or between genetic similarity and habitat characteristics (r = 0·019; P = 0·356).
Most of the variation (51·9 %) in phenotypic traits (Table 3) of Q. parviflora could be explained by two axes in the PCA. Sclerophylly, C : N ratio, concentration of tannins and sulphur in leaves showed a direct relationship with axis 1, while contents of N, P, K and Na correlated negatively with this axis. Concentration of Ca, Al, Mn and leaf area had a direct relationship with axis 2, while the water status had a negative relationship with this axis (Fig. 2). Individuals from cerradão were plotted on the left side of axis 1, which separated most individuals in this population from those in cerrado sensu stricto and campo sujo, which were plotted mainly on the right side of axis 1 (Fig. 2).
|
|
Most of the variation in habitat characteristics (Table 4) could be explained by two axes (66·8 %). Most of the habitat characteristics were correlated with axis 1—potential acidity, aluminium saturation, total and effective capacities of cation exchanges, calcium, magnesium, potassium, phosphorous, organic matter, manganese and water in soil (Fig. 3). Among these variables, only aluminium saturation showed a negative correlation. Water in soil and K were positively correlated with the second axis, whereas Si and N-nitrate were negatively correlated. Individual plants from cerrado sensu stricto and cerradão could be separate by axis 1, while axis 2 separated the individuals from campo sujo from the other physiognomies (Fig. 3). There was no significant correlation between phenotypic traits and habitat characteristics (r = 0·012; P = 0·594).
|
|
A significant correlation obtained between axis 1 from phenotype PCA and axis 2 from habitat PCA (r = 0·70; P < 0·0001) indicated that some traits of Q. parviflora trees, such as sclerophylly, C : N ratio, and N, P, K and Na concentrations in leaves, are related to features of the habitat. The other axis from phenotype, genotype and habitat PCA did not show any significant relationships (in all P > 0·05).
Herbivory and effects of the genetic variability, phenotypic and habitat traits
Adult insects of 79 species, in six orders, were found on leaves of Q. parviflora. Phytophagy was the most common diet among insects in the three types of vegetation (n = 64 species).
Variation in number of herbivorous insect species and attack by insects on leaves and shoots of Q. parviflora (Table 5) could be partially explained by some of the axes obtained by PCA based on genetic variability, phenotype and habitat characteristics (Table 6). Number of insect species was best explained by habitat characteristics (partial r2 with HabAxis 2 = 0·24; P = 0·006), such as concentration of silica, nitrogen and water in the soil (Table 6). In addition, variation in percentage of damaged leaves (free-living herbivore attack) was better explained by genetic variability (PCA axes 2 and 4) and habitat (PCA axis 2). On the other hand, phenotypic traits were the most important factor affecting both variations in percentage of leaves with galls (partial r2 with PhenAxis 1 = 0·14; P = 0·04), and number of galls per leaf (partial r2 with PhenAxis 2 = 0·18; P = 0·019) (Table 6). Variation in the total number of galls per plant could not be explained by any of the axes obtained in PCA.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
For RAPD markers, there is a high genetic diversity and low genetic divergence among Q. parviflora individuals from the three types of vegetation of the cerrado at the Ecological Station of Pirapitinga. A large percentage of variation accounting for differences among individuals within populations is commonly found in outcrossing plant species (e.g. Dawson and Powell, 1999
; Gurgerli et al., 1999; Hsiao and Lee, 1999; Jordano and Godoy, 2000; Gomes, 2001). Thus, these results were expected due to the exclusive sexual reproduction, and the alogamous and self-incompatible reproductive system, in Q. parviflora (Barbosa, 1983). Although PCA analysis based on phenotype and habitat characteristics could separate most of the individuals from the three sites, there was no significant relationship between genetic similarity, phenotype and habitat characteristics. In fact, genetic similarity among individuals from the three types of vegetation indicated that gene flow between sites is high and they probably belong to the same population. Variation in habitat quality among physiognomies may not be sufficient to cause genetic divergence due to differential selective pressure when there is no barrier to gene flow between sites.
There were no significant relationships between genetic similarity and the chemical and physical characteristics of plants, and between genetic similarity and the habitat traits of Q. parviflora. No significant correlation was found between phenotype traits and habitat characteristics. Telles et al. (2001) found a significant relationship between plant and soil characteristics for Eugenia dysenterica (Myrtaceae). They pointed out that the geographical distribution would be the main factor affecting the genetic divergence among local populations of this species. In addition, they did not find any significant correlation between genetic distance and plant phenotype, similar to the results presented here.
Habitat traits explained the variation in the number of herbivorous insect species on Q. parviflora better. The availability of water and nutrients in the soil and weather conditions are important factors that affect the growth and quality of plants (White, 1969, 1984), and indirectly influence the diet, and thus type, abundance and growth of herbivorous insects. The increase in soil nutrient concentration influences the plant's physiology and changes the nutrient composition, such as protein (Schoonhoven et al., 1998), and secondary metabolite concentrations (Gershenzon, 1984). Differences in the number of insect herbivores between the three types of vegetation were found, with most herbivorous species occurring on plants of the campo sujo and fewest on plants in the cerradão (S. J. Gonçalves-Alvim and G. W. Fernandes, unpublished results).
Variation in percent of damaged leaves was best explained by genetic variability and habitat. Similarly, Ribeiro and Brown (1999) observed a significant relationship between leaf damage by herbivorous insects and the genetic dissimilarity in two species of Tabebuia (Bignoniaceae). Also, an association was found between increasing concentrations of tannins and sclerophylly, and the increase in percentage of leaf area lost by herbivore attack.
Contrary to the results of some authors (e.g. Coley, 1983; Salatino, 1993; Turner, 1994; Ribeiro et al., 1998), sclerophylly may not play an import role in protecting Q. parviflora against herbivores, indicating that the insects could be adapted to the sclerophyllous characteristics of this species. Ribeiro and Brown (1999) did not observe any correlation between severity of herbivory and toughness of leaves in T. aurea and T. ochracea (Bignoniaceae). The results presented here also differ from those obtained by Marquis et al. (2001), who found a negative correlation of the concentrations of secondary compounds and their capacity to condense proteins.
The results given here suggest that gall-forming insects respond to the nutritional quality of Q. parviflora because leaves richer in nutrients and poorer in tannins had more galls. Also, these results do not corroborate the plant stress hypothesis (White, 1969, 1976, 1984; Rhoades and Cates, 1976; Rhoades, 1979; Mattson and Haack, 1987), which predicts that nutritionally stressed plants should have smaller concentrations of chemical defences, so becoming more susceptible to attack by herbivores. In addition, gall-forming insects were more abundant in plants on more fertile soils and with less sclerophyllous leaves, again not corroborating the soil fertility hypothesis (Fernandes and Price, 1988, 1991; Blanche and Westoby, 1995) that predicts higher richness and abundance of gall-forming insects in environments with smaller concentrations of nutrients in the soil.
In conclusion, the increase of Si and N and decrease of K and water in soil were positively related with the percentage of damaged leaves, leaf area lost and number of insect species on Q. parviflora. Additionally, the increasing concentrations of tannins, S, C : N ratio and sclerophylly were associated with increased percentage of leaf area lost by herbivore attack. A decrease in sclerophylly, concentration of tannins and C : N ratio, together with increasing concentrations of nutrients, such as N, P, K, Ca and Na, increase attack by gall-forming insects. Thus, in Q. parviflora the indirect effects of nutrient concentrations in soil (i.e. nutrients available to the plant) and the direct effects of defensive compounds (i.e. tannins) could influence the behaviour of herbivorous insects. However, free-feeding herbivorous insects may avoid the negative effects of tannins in their feeding and increase consumption of leaves to satisfy their nutritional requirements by moving from one host plant to another. Gall-forming insects, on the other hand, are more affected by the quality of leaves of Q. parviflora, with fewer sclerophyllous leaves that have a larger concentration of nutrients and a smaller concentration of tannins being attacked more by gall-forming insects.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
We thank A. Alvim Filho, B. Dourado, F. Silveira, V. Freire and T. Lana for field assistance, Dr Karsten Schonrogge and an anonymous reviewer for very helpful comments that significantly improved this manuscript. Support by the Sociedade Brasileira de Cultura Japonesa/Fundo Bunka de Pesquisa-Banco Sumitomo, WWF-(CSR 174-00), USAID, US Fish and Wildlife Service is gratefully acknowledged. Logistical support for plant sampling was provided by ESP and Instituto Brasileiro de Recursos Renováveis e Meio Ambiente (IBAMA). Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA), Universidade Federal de Viçosa, Universidade Federal de Uberlândia and Universidade Católica de Brasília provided plant and soil analyses. This study was in partial fulfilment of the requirements for the PhD degree of S.J.G.-A. at Universidade Federal de Minas Gerais for which a scholarship was provided by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Anderson SS, McCrea KD, Abrahamson WG, Hartzel LM. 1989. Host genotype choice by the ball gallmaker Eurosta solidaginis (Diptera: Thephritidae). Ecology 70: 1048–1054.[CrossRef]
Barbosa AAA. 1983. Aspectos de ecologia reprodutiva de três espécies de Qualea (Vochysiaceae) num cerrado de Brasília-DF. MSc Thesis, Universidade de Brasília, Brasília.
Blanche KR, Westoby M. 1995. Gall-forming insect diversity is linked to soil fertility via host plant taxon. Ecology 76: 2334–2337.[CrossRef]
Bazzaz FA, Chiarello NR, Coley PD, Pitelka LF. 1987. Allocating resources to reproduction and defense. Bioscience 37: 58–67.
Carlson RM, Cabrera RI, Paul JL, Quick J, Evans RY. 1990. Rapid direct determination of ammonium and nitrate in soil and plant tissue extracts. Communication on Soil Science and Plant Analyses 21: 1519–1529.
Chapin FS, Bloom AJ, Field CB, Waring RH. 1987. Plant response to multiple environmental factors. Bioscience 37: 49–57.
Coley PD. 1983. Intraspecific variation in herbivory on two tropical tree species. Ecology 64: 426–433.[CrossRef]
Coley PD, Barone JA. 1996. Herbivory and plant defenses in tropical forests. Annual Review of Ecology and Systematics 27: 305–335.[CrossRef][Web of Science]
Collevatti RG, Schoereder JH. 1995. Microclimate ordination and litter arthropod distribution. Ciência e Cultura 47: 38–41.
Cornelissen T, Fernandes GW. 2001. Defense, growth, and nutrient allocation in the tropical shrub Bauhinia brevipes (Leguminosae). Austral Ecology 26: 246–253.[CrossRef]
Dawson JK, Powell W. 1999. Genetic variation in the Afromontane tree Prunus africana, an endangered medicinal species. Molecular Ecology 8: 151–156.
Doyle JJ, Doyle JL. 1987. Isolation of plant DNA from fresh tissue. Focus 12: 13–15.
Eiten G. 1972. The cerrado vegetation of Brazil. Botanical Review 38: 201–341.
Elliott CL, Snyder GH. 1991. Autoclave-induced digestion for the colorimetric determination of silicon in rice straw. Journal of Agricultural Food and Chemistry 39: 1118–1119.[CrossRef]
Ernest, KA. 1989. Insect herbivory on a tropical understory tree: effects of leaf age and habitat. Biotropica 21: 194–199.[CrossRef][Web of Science]
Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479–491.[Abstract]
Felfili JM, Silva MC Jr. 1993. A comparative study of cerrado (sensu stricto) vegetation in Central Brazil. Journal of Tropical Ecology 9: 277–289[Web of Science]
Felippe GM, Dale JE. 1990. The effects of phosphate supply on growth of plants from the Brazilian Cerrado: experiments with seedlings of the annual weed, Bidens gardneri Baker (Compositae) and the tree, Qualea grandiflora (Mart.) (Vochysiaceae). Oecologia 82: 81–86.[CrossRef]
Fernandes GW, Price PW. 1988. Biogeographical gradients in galling species richness. Tests of hypotheses. Oecologia 76: 161–167.[CrossRef][Web of Science]
Fernandes GW, Price PW. 1991. Comparisons of tropical and temperate galling species richness: the roles of environmental harshness and plant nutrient status. In: Price PW, Lewinsohn TM, Fernandes GW, Benson WW, eds. Plant-animal interactions: evolutionary ecology in tropical and temperate regions. New York: John Wiley and Sons, 91–115.
Fritz RS. 1990. Effects of genetic and environmental variation on resistance of willow to sawflies. Oecologia 82: 325–332.[CrossRef]
Gershenzon J. 1984. Changes in the levels of plant secondary metabolites and water and nutrient stress. Recent Advances in Phytochemistry 18: 273–320.
Gomes V. 2001. Distribuição intra e interpopulacional da variablidade genética da espécie endêmica e ameaçada Baccharis concinna (Asteraceae). MSc Thesis. Universidade Federal de Minas Gerais, Belo Horizonte.
Gurgerli F, Eichenberger K, Schneller JJ. 1999. Promiscuity in populations of the cushion plant Saxifraga oppositifolia in the Swiss Alps as inferred from random amplified polymorphic DNA (RAPD). Molecular Ecology 8: 453–461.[CrossRef]
Hallmark C, Wilding L, Smeck N. 1982. Silicon. Methods of soil analysis. Part 2. Chemical and microbiological properties. Agronomy Monograph 9: 263–273.
Hagerman AE. 1987. Radial diffusion method for determining tannin in plant extracts. Journal of Chemical Ecology 13: 437–449.[CrossRef][Web of Science]
Haridasan M. 1982. Aluminum accumulation by some cerrado native species of central Brazil. Plant and Soil 65: 265–273.[CrossRef]
Hartnett DC, Abrahamson WG. 1979. The effects of stem gall insects on life history patterns in Solidago canadensis. Ecology 60: 910–917.[CrossRef][Web of Science]
Hsiao JY, Lee SM. 1999. Genetic diversity and microgeographic differentiation of Yushan cane (Yushamnia niitakayamensis; Poaceae) in Tawain. Molecular Ecology 8: 263–270.
Johnson CM, Ulrich A. 1959. Analytical methods for use in plant analysis. Bulletin 766. Berkeley: University of California, Agricultural Experiment Station.
Jordano P, Godoy JA. 2000. RAPD variation and population genetic structure in Prunus mahaleb (Rosaceae), an animal-dispersed tree. Molecular Ecology 9: 1293–1305.[CrossRef][Medline]
Karban R. 1992. Plant variation: its effects on populations of herbivorous insects. In: Fritz RS, Simms EL, eds. Plant resistance to herbivores and pathogens: ecology, evolution and genetics. Chicago: University of Chicago Press, 195–215.
Kempers AY, Zweers A. 1986. Ammonium determination in soil extracts by the salicylate method. Soil Science and Plant Analyses 17: 715–723.
Mantel NA. 1967. The detection of disease clustering and generalized regression approach. Cancer Research 27: 209–220.
Marquis RJ, Diniz IR, Morais HC. 2001. Patterns and correlates of interspecific variation in foliar insect herbivory and pathogen attack in Brazilian cerrado. Journal of Tropical Ecology 17: 127–148.[CrossRef][Web of Science]
Mattson WJ, Haack RA. 1987. The role of drought stress in provoking outbreaks of phytophagous insects. In: Barbosa P, Schultz JC, eds. Insect outbreaks: ecological and evolutionary perspectives. Orlando, FL: Academic Press, 365–394.
Meyer GTA, Keliher PN. 1992. An overview of analysis by inductively coupled plasma-atomic emission spectrometry. In: Montaser A, Golightly, DW, eds. Inductively coupled plasmas in analytical atomic spectrometry. New York: VCH Publishers, 473–505.
Moles AT, Westoby M. 2000. Do small leaves expand faster than large leaves, and do shorter expansion times reduce herbivory? Oikos 90: 517–524.[CrossRef][Web of Science]
Mullin JB, Riley JP. 1955. The colorimetric determination of silicate with reference to sea and natural waters. Analitical Chemichal Acta 12: 162–176.[CrossRef]
Mutikainen P, Walls M, Ovaska J, Keinanen M, Julkunem-Tiitto R, Vapaavuori H. 2000. Herbivore resistance in Betula pendula: effect of fertilization, defoliation, and plant genotype. Ecology 81: 49–65.[CrossRef][Web of Science]
Myers JH. 1985. Effect of physiological condition of the host plant on the ovipositional choice of the cabbage white butterfly, Pieris rapae. Journal of Animal Ecology 54: 193–204.[CrossRef]
NTSYS. 1997. Version 2.02. Applied Biostatistics Inc., EUA.
Rhoades DF. 1979. Evolution of plant chemical defense against herbivores. In: Rosenthal AG, Janzen DH, eds. Herbivores: their interactions with secondary plant metabolites. New York: Academic Press, 3–54.
Rhoades DF, Cates RG. 1976. Toward a general theory of plant antiherbivore chemistry. In: Wallace J, Mansell RL, eds. Recent advances in phytochemistry, Vol. 10. New York: Plenum Press, 168–213.
Ribeiro SP, Brown VK. 1999. Insect herbivory in tree crowns of Tabebuia aurens and T. ochracea (Bignoniaceae) in Brazil: contrasting the Cerrado with the Pantanal Matogrossense. Selbyana 20: 159–170.
Ribeiro SP, Carneiro MAA, Fernandes GW. 1998. Free-feeding herbivores along environmental gradients in Serra do Cipó: basis for a management plan. Journal of Insect Conservation 2: 1–12.[CrossRef]
Ribeiro SP, Pimenta HR, Fernandes, GW 1994. Herbivory by chewing and sucking insects on Tabebuia ochracea. Biotropica 26: 302–307.[CrossRef]
Salatino A. 1993. Chemical ecology and the theory of oligotrophic scleromorphism. Anais da Academia Brasileira de Ciências 65: 1–13.
Scion-Image software. 1998. Scion Corporate. Freely available in http://www.scioncorp.com/
Schneider S, Roesslim D, Excoffier L. 2000. Arlequin 2000: a software for population genetics data analysis. Genetics and Biometry Laboratory, University of Genova, Switzerland.
Scholander PF, Bradstreet ED, Hammel HT, Hemmingsen EA. 1965. Sap pressure in vascular plants. Science 148: 339–346.
Scholander PF, Hammel HT, Hemmingsen EA, Bradstreet ED. 1964. Hydrostatic pressure and osmotic potential of mangrove and some other plants. Proceedings of Naturalist Academy of Science 52: 112–119.
Schoonhoven LM, Jermy T, Von Loon JJA. 1998. Insect-plant biology: from physiology to evolution. London: Chapman Hall.
Silva FC. 1999. Manual de análises químicas de solos, plantas e fertilizantes. Brasília: Empresa Brasileira de Pesquisa Agropecuária.
Sneath PH, Sokal RR. 1973. Numerical taxonomy. New York: Freeman.
Stein SJ, Price PW. 1995. Relative effects of plant resistance and natural enemies by plant developmental age on sawfly (Hymenoptera: Tenthredinidae) preference and performance. Environmental Entomology 24: 909–916.
Swain T. 1979. Tannins and lignins. In: Rosenthal GA, Janzen DH, eds. Herbivores: their interactions with secondary plant metabolites. New York: Academic Press, 657–682.
Telles MPC, Silva RSM, Chaves LJ, Coelho ASG, Diniz Filho JAF. 2001. Divergência entre subpopulações de caigateira (Eugenia dysenterica) em reposta a padrões edáficos e distribuição espacial. Pesquisa Agropecuária Brasileira 36: 1387–1394.
Turner IM. 1994. Sclerophylly: primary protective? Functional Ecology 8: 669–675.[CrossRef][Web of Science]
Walkely A, Black IA. 1934. An examination of Degtjareff method for determining soil organic matter and a proposed modification of the cromic acid titration method. Soil Science 37: 29–37.[Web of Science]
White TCR. 1969. An index to measure weather-induced stress of trees associated with outbreaks of psyllids in Australia. Ecology 50: 905–909.[CrossRef]
White TCR. 1976. Weather, food and plagues of locusts. Oecologia 22: 119–134.[CrossRef][Web of Science]
White TCR. 1984. The abundance of invertebrate herbivores in relation to the availability of nitrogen in stressed food plants. Oecologia 63: 90–105.[CrossRef][Web of Science]
Whittaker K, Feeny P. 1971. Allelochemics: chemical interactions between species. Science 171: 757–769.
Woodman RL, Fernandes GW. 1991. Differential mechanical defense: herbivory, evapotranspiration, and leaf hairs. Oikos 60: 11–19.[CrossRef][Web of Science]
Zangerl A, Berenbaum MR. 1997. Cost of chemically defending seeds: furanocoumarins and Pastinaca sativa. American Naturalist 150: 491–504.[CrossRef][Web of Science]
Zar JH. 1996. Biostatistical analysis. Englewood Cliffs, NJ: Prentice Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||