Introduction

Life-history parameters of tetranychid mite species are controlled by intrinsic and extrinsic factors, being host plant conditions among the most important (Huffaker et al. 1969; Wrensch 1985). Crooker (1985) emphasized that morphological and chemical plant features are thought to be associated with host plant resistance, inducing differences in mite development rate, fecundity, longevity and, consequently, in mite fitness. More recent research indicated that differential responses in phytophages can be raised by host plant quality, which also depends on quantity and nature of primary and secondary metabolites (Awmack and Leather 2002; van den Boom et al. 2003). These factors may vary among plant species and even among cultivars. As an example, fecundity of Tetranychus cinnabarinus (Boisduval) varied from 19 to 400 eggs in several Cucurbita and Lagenaria accessions (Edelstein et al. 2000), while Tetranychus urticae Koch showed greater population increases in strawberry ‘Hapil’ and ‘Pegasus’ cultivars than in others (Sonneveld et al. 1997). Furthermore, morpho-anatomical features such as trichomes and foliar glands, and epidermis-cuticle strata thickness can constitute physical barriers for tetranychid feeding (e.g. in cotton; Bailey et al. 1978).

The avocado brown mite, Oligonychus punicae (Hirst), is considered an important tetranychid pest in southern California (USA), as high population densities can cause severe defoliation on several avocado cultivars (McMurtry and Johnson 1966; McMurtry 1985). This tetranychid mite predominantly feeds on the upper leaf surface, although feeding could extend to the lower leaf surface at high population levels (Tomczyk and Kropczynska 1985). Damage in host plants is shown by a bronze tone on the leaves, which was shown to be associated with the rates of oviposition and female production (McMurtry 1970).

In tropical America, the avocado brown mite has been reported in more than 20 plant species such as Mangifera indica L., Musa sapientum L., Punica granatum L. and Vitis vinifera L. (Ochoa et al. 1994; Bolland et al. 1998). In Venezuela, it was recorded on Musa spp. from Sur del Lago, Zulia State (Freitez and Alvarado 1978; Quirós 1978). More recently it has been observed as an occasional pest on grape from Lara State where its feeding delays fruit ripening. This study was undertaken to investigate how feeding on leaves from six grape cultivars from Lara State influences life-history parameters of the avocado brown mite.

Material and methods

Mite collection and maintenance

Oligonychus punicae was collected on infested grape leaves in a vineyard of the Criolla Negra cultivar, located at Instituto de la Uva (IUVA) (9°48′ N, 69°47′ W), belonging to Universidad Centroccidentral ‘Lisandro Alvarado’, El Tocuyo, Morán Municipality, Lara State, Venezuela. In the laboratory, the mites were cultured on Criolla Negra vines for 6 months before the start of experiments. Experiments were conducted in a room at 27 ± 2°C, 80 ± 10% RH, and a photoperiod of 12:12 h.

Estimation of O. punicae developmental time

Oligonychus punicae development was determined when feeding on leaves from Tucupita, Villanueva, Red Globe, Sirah, Sauvignon and Chenin Blanc grape cultivars, grown at El Tocuyo. Tucupita, Villanueva and Red Globe cultivars were sown under headlike arbor, while Sirah, Sauvignon and Chenin Blanc were conducted in trellis system. Studies on biological features were carried out on mature leaves sampled from the middle part of the stem of >5-year-old plants.

Ten adult females and three males were randomly selected from the laboratory culture and placed together on each of 10 rearing units containing grapevine leaf discs (3 cm diameter) from each cultivar, as described by Helle and Overmeer (1985), to lay eggs. Mites were removed 24 h later. The egg cohorts were observed at 12-h intervals. After that, a 2-day-old female and a male were both placed on a leaf disc from each of the six grape cultivars. Once one or three eggs were laid per leaf disc, the male and female mites were removed. Observations were made at 12-h intervals to determine egg hatching time and developmental time of larval, protonymphal and deutonymphal stages. Ten replications were run.

Estimation of O. punicae reproductive parameters and longevity

Pre-oviposition, oviposition and post-oviposition periods, total fecundity, daily oviposition rate and female longevity were determined on each of the six grape cultivars. A newly emerged female and one adult male mite were placed together on leaf discs (3 cm diameter) from each of the grape cultivars as described above. Pre-oviposition period was recorded at 6-h intervals, while other reproductive parameters and longevity were recorded every 12 h. Leaf discs were removed and replaced by new ones at 4-day intervals. Ten replications were run.

Life table parameters

Net reproduction rate (R 0), mean generation time (T), intrinsic rate of increase (r m) and finite rate of increase (λ) of O. punicae were calculated from data obtained for development time and reproductive parameters according to Birch (1948).

Estimation of chemical and anatomical features on grape cultivars

Two experiments were conducted to determine the content of secondary metabolites and some anatomy features of leaves of the six grape cultivars in order to elucidate their possible influence on O. punicae life-history parameters. In experiment 1, samples of grape leaves from each cultivar were collected at IUVA vineyards under the following two conditions: (a) mite-infested leaves (MIL), which were naturally infested with no chemical applications included, and (b) non-infested leaves (NIL) which were frequently sprayed with water to prevent mite population increase. Fully expanded leaves were taken from the middle third of the primary shoot, and in the case of MIL just leaves showing more than 50% of O. punicae damage symptoms were collected. Both MIL and NIL samples were dried at room conditions and then individually ground until obtaining a fine powder from each cultivar. Five hundred milligram of the leaf powder obtained from 25 to 30 leaf discs (3 cm diameter) from each cultivar were mixed with 96% alcohol during 24 h to obtain an ethanolic extract by using a rotoevaporator BrinkmamMR (Mod. RE111), at 100°C.

Secondary metabolites were determined in ethanolic extract as described by Marcano and Hasegawa (2002). Alkaloids, flavonoids, tannins and polyphenols were separated and quantified by thin-layer chromatography (Cromatoplates MerckMR to silica/gel 60 F254 TLC, 0.25 mm thin, 6.5 × 2.5 cm). One ethanolic extract drop (7 μl) from each cultivar was added to cromatoplates and treated with a 9:1 water/acetic acid solution, to separate tannins and polyphenols which were identified by adding 0.1% FeCl. Alkaloids and flavonoids were separated using hexane as a solvent and printings were observed under UV light as described by Marcano and Hasegawa (2002). Bands containing isolated secondary compounds were cut and weighed (silica/gel + secondary metabolite). A similar area was obtained from a clean silica/gel cromatoplate (silica/gel) and weighed for comparison purposes. The difference in weights indicates the amount of the particular secondary plant compound (in 7 μl extract). Based on these weights the contents (per g leaf material) could be calculated in a straightforward way.

In experiment 2, three plants of each of the six cultivars were selected randomly from the vineyard to determine the cuticle-epidermis thickness and trichome density on the adaxial leaf surface. Sub-samples of three leaves per plant were randomly chosen and fixed in FAA medium (5% formaldehyde, 5% acetic acid, 90% alcohol). Subsequently, the sub-samples were divided in two groups. In group 1, leaves were subjected to mesophyll digestion with Jeffrey liquid (Johansen 1940) in order to detect leaf trichome presence and density. In group 2, cross-sections were obtained using a rotary microtome (Reichert-Jung) in order to measure cuticle-epidermis strata. To better show leaf anatomy, previous to microscope observation, leaves from both groups were stained with toluidine as described by Montenegro and Gómez (1997).

Statistical analysis

Biological parameters were determined in a completely randomized design with time-repeated measurements, analyzed through one-way ANOVA and Tukey tests, to separate treatment means. Secondary metabolite data were analyzed by one-way ANOVA and correlated to life-history parameters to determine potential association between them. Anatomy data were subjected to ANOVA and Tukey tests to determine differences between cultivars. All analyses were performed by using Statistix software version 8.0.

Results and discussion

Influence of grape cultivars on O. punicae developmental time

The total developmental time (egg-adult) of O. punicae showed significant differences among grape cultivars (P = 0.004, df = 5, F = 4.78). However, multiple linear regression analysis showed no relation between total developmental time and alkaloids, flavonoids, tannins or polyphenol content levels of the cultivars (P = 0.175, F = 1.86, R 2 = 0.12). On Tucupita and Sauvignon, mites spent 8.2 and 8.3 days to reach adulthood, while on Villanueva, Red Globe, Chenin Blanc and Sirah development was ca. 7–11% slower (Table 1). Previous studies have demonstrated cultivar effects on life cycle and population increase of a variety of tetranychid species. Opit et al. (2001), for example, observed that T. urticae populations increased faster on geranium cultivar ‘Amethyst’ than on ‘Sybil Holmes’ under glasshouse conditions. They hypothesized that possibly higher phosphorous content in ‘Amethyst’ caused the population increase since it has been demonstrated previously that phosphorous may enhance T. urticae fecundity. Another example is the developmental time of Amphitetranychus viennensis (Zacher), which was found to be lower when reared on apple cultivars ‘Starkrimson Delicious’ and ‘Golden Delicious’ (10.7 days) than on ‘Amasya’ and ‘Starking Delicious’ (11.7 days) (Kasap 2003).

Table 1 Developmental time of different life stages of O. punicae in various grapevine cultivars
Full size table

Foliar anatomy of the grape cultivars seemed not to hinder O. punicae feeding, because mites reached adulthood also on cultivars Red Globe and Sauvignon, despite the greater cuticle-epidermis thickness (Table 2). Perhaps this need not come as a surprise—the length of the mites’ stylet, measured from basis to apex is approximately 43 ± 2.2, 74 ± 7.5 and 110 ± 4.0 μm in larvae (N = 15), males (N = 7) and females (N = 11), respectively, which is longer than even the thickest cuticle-epidermis measured, so probably mites are able to feed on the lower leaf strata on all cultivars (Table 2).

Table 2 Strata cuticle-epidermis thickness in several grapevine cultivars
Full size table

Studies about the influence of leaf anatomy on tetranychid life-history parameters are not conclusive. Skorupska (1998) showed that stoma count on the abaxial surface, spongy and palisade mesophyll and total leaf lamina width affected growth of A. viennensis in apple cultivars. Conversely, Nukenine et al. (2000) did not find any relation between anatomical features in cassava and resistance to the cassava green mite.

Reproductive parameters and longevity

Periods of pre-oviposition, oviposition and post-oviposition in O. punicae females varied among grape cultivars. Shorter pre- and post-ovipositional periods were found on Chenin Blanc leaves (1.2 and 0.9 days, respectively), while on others cultivars pre- and post-oviposition periods ranged from 1.6–2.0 to 1.2–2.5 days, respectively (Table 3). Oviposition periods lasted from 6.7 days (on Villanueva) to 16.1 days (on Sauvignon). In previous studies O. punicae had a pre-ovipositional period of 1.3 days and an oviposition period of 13.3 days (Tanigoshi and McMurtry 1977) or 2.0 days (McMurtry and Johnson 1966) on avocado trees.

Table 3 Adult longevity, pre-ovipostion, oviposition and post-oviposition times (days) and fecundity (eggs/female/day) of O. punicae on several grapevine cultivars
Full size table

Average daily egg production was highest on Tucupita (2.8 eggs/female/day) and lowest on Sirah (0.9 eggs/female/day) (Table 3). Daily oviposition rate ranged from 2.0 to 6.1 eggs/female up to day 7 in females feeding on Tucupita leaves, while on Chenin Blanc, Red Globe, Sauvignon, Sirah and Villanueva leaves it varied from 1.1–6.1, 0.2–5.3, 0.3–5.0, 0.2–2.7 to 0.5–3.5 eggs/female, respectively (Fig. 1). Daily oviposition rates in this study were higher than reported by McMurtry and Johnson (1966) or Tanigoshi and McMurtry (1977) for O. punicae females reared on avocado leaves.

Fig. 1
figure 1

Daily oviposition rate of O. punicae females feeding on grapevine leaves cultivars Red Globe (a), Chenin Blanc (b), Sauvignon (c), Tucupita (d), Sirah (e) and Villanueva (f)

Full size image

Also, female longevity of O. punicae was affected by grape cultivar: females lived longest on Sauvignon (17.5 days), and shortest on Villanueva (8.1 days) (Table 3). Similarly, Kerguelen and Hoddle (2000) found that mean female O. perseae longevity increased by 100% from 12 days in May to 24 days in July in the susceptible avocado cultivar Hass, while in more resistant cultivars, Pinkerton and Lamb Hass, longevity dropped by 30% over the same period.

An effect of host plant on reproduction has been established for several Tetranychid species (e.g., de Ponti 1977; Ribeiro et al. 1988; Hilker and Meiners 2002; Praslička and Huszár 2004). In our findings, reproductive parameters of O. punicae seemed to be associated with flavonoid content of grape cultivars: the higher the flavonoids content, the lower the mites’ fecundity (R 2 = 63.5, F = 23.48, P < 0.001) (Fig. 2). Previous studies have demonstrated that grape leaves and fruits synthesize phenolic compounds in response to fungal attacks or abiotic factors (Morrissey and Osbourn 1999). Furthermore, Harborne (1994) hypothesized that low molecular weight phenol compounds could act synergistically with tannins to provide plant resistance.

Fig. 2
figure 2

Lineal regression between flavonoid content and mean oviposition of O. punicae on grapevine cultivars (y = 4.78 − 36.42x)

Full size image

Life table parameters

The net reproductive rate (R 0) was highest on Sauvignon and lowest on Villanueva (Table 4). The intrinsic rate of increase (r m) ranged from 0.31 (on Chenin Blanc) to 0.14 (on Villanueva). Based on our life-history parameters, the cultivars Sirah and Villanueva proved to be the least suitable hosts for this pest mite, due to slower development and lower fecundity, longevity and r m-values. Sauvignon and Tucupita were much better hosts, based on the higher mite reproduction under laboratory conditions. Further field studies should be carried out to determine the interaction between plant, pest and environment in Lara state vineyards in order to establish more efficient pest management strategies.

Table 4 Life table parameters of O. punicae feeding on six grapevine cultivars
Full size table