STUDIO DELLE MODALITÀ DI ALIMENTAZIONE DI STEPHANITIS PYRI E DEL COMPORTAMENTO ALIMENTARE DELL'INSETTO SU DIFFERENTI VARIETÀ DI PERO

Stephanitis pyri (Fabricius) (Heteroptera Tingidae), also known as “pear lace bug”, is a heavy fruit pest of Rosaceae common in Southern Europe. Its preferred host is Pyrus communis, a typical agricultural crop tree of high economical relevance in the region Emilia-Romagna (Northern Italy). The damages inflicted to leaves and fruits are marked yellowing and vanish-like tar spots, resulting into complete decline of untreated plants. The number of studies on Tingidae family is still limited, although several species are economically relevant in regions where pear tree orchards are widespread. A study to collect more information on behaviour and especially on feeding habits of the most relevant Tingidae species, S. pyri, was therefore initiated in 2006. The purpose of the study was to observe and record for the first time the feeding behaviour of S. pyri by Electrical Penetration Graph (EPG), to identify and analyze the different feeding phases and the interactions between the insect and the plant during the feeding process. The identification of EPG phases is based on waveforms, different in number and duration but typical of each species. To collect more information on feeding behaviour, the stylets of the insects were cut (stylectomy) during feeding phases and, after resin inclusion, the plant tissues containing the stylets were examined by optical and electron microscopy to match each stylet position to the typical waveform detected. These data allow to identify the plant tissues involved in feeding, establishing the degree of food specialization of the pest. Comparisons of duration and other parameters of feeding phases were then performed in different hosts, such as the pear varieties “William” and “Conference”, economically relevant cultivars, typical of Emilia Romagna. A study on feeding behaviour during host pre-contact and pre-penetration phases was also performed by choice tests, both in the field and in laboratory. Field experiments were also performed on S. pyri adult and nymph populations, to detect possible differences in development related the genotypes of the pear cultivars analyzed. The leaves of the two varieties are actually different from a morphological point of view: in Conference the leaves are more bended and often overlapping (thus offering a better hiding for insects), while in William they are flat and well separated. To exclude morphological factors in host choice, the experiments were repeated in laboratory on flat, round leaf disks. For field experiments, greenhouses with anti-aphid nets were prepared to restrain each individual S. pyri introduced and born, thus allowing a continuous time recording of insects and of their distribution on plants, and preventing the introduction of other species. The setting of laboratory conditions similar to the orchard ones allowed observations of typical features of behaviour and life cycle, such as orienteering, feeding choices, plant sensitivity or resistance to the pest, longevity, fitness, egg laying habits, etc. Besides choice tests, also no-choice tests were performed on the two pear cultivars to collect more information on development of S. pyri adult and nymph populations. The two cultivars were independently tested to avoid interferences in the host choice. The data collected could also be useful to increase information on the pear cultivar tested, to identify interesting traits for agronomical practices with low environmental impact. MATERIALS AND METHODS Insects Adult male and female individuals of Stephanitis pyri, progeny of individuals collected in orchards between the provinces of Ferrara and Rovigo (Northern Italy), were employed in all experiments. The parent individuals were collected by an entomological sheet on trees of wild pear, Pyrus pyraster (Burgsd.), common pear, Pyrus communis (L.) and apple, Malus pumila (Mill.). The insects were adult males and females of first, second and third generation, raised on P. communis cultivar “Abate” in full vegetative stage, in temperature-controlled rooms at 22 °C and 16:8 hours (day/night) photoperiod. Only for the choice test experiments on round-cut leaf portions, first generation S. pyri adults raised on Conference were employed, since this cultivar is considered more sensitive to the pest attack in comparison to William. Plants The plants employed in all experiments were one-year grafted plants of Pyrus communis belonging to the William, Abate and Conference cultivars, respectively grafted to Butyrra Hardi / BA29 (William and Abate) and on quince Sydo (Conference). All plants were free from chemical pest and fungus disease treatments, raised in plastic pots with standard, regularly watered soil for tree culture. Feeding habits and stylet position Technical equipments Many instruments were simultaneously employed: the Electrical Penetration Graph (EPG-DC, Wageningen University), a software for data reading (PROBE 3), an electric microknife (RF microcautery unit), a Faraday cage, a micromanipulator Narishige (Japan), a binocular Nikon 48161 microscope, an optical microscope Nikon Eclipse E200 (Tokyo, Japan), a Photoshop programs for image analysis, a transmission electron microscope (TEM) Hitachi H-800 (Electronic Microscopy Centre of the University of Ferrara), an ultramicrotome Reichart Om U2, (Germany). Semithin sections (about 1 μm thick) were observed at the optical microscope after staining with methylene blue; ultrathin sections (600 Å) for TEM were contrasted with uranyl acetate and lead citrate. EPG-DC parameters A comparison was performed between the cultivars Conference and William. For each 8-hr EPG-DC test, a different insect and a different plant were employed: all insects were placed on plants in standardized positions. A total of 20 trials with 20 different insects were performed for each cultivar. All results were validated, considering only those in which no interruptions were recorded in the trial period. The statistical analysis of all EPG parameters and the data obtained by comparison of the insect feeding behaviour in the two cultivars was performed by the Mann-Whitney test (p<0,05). Choice tests Choice test on leaves in laboratory. A total of 492 male and female S. pyri IIIrd generation adults were employed, raised on Abate. Each group of 6 individuals was placed inside a 15-cm Petri dish containing 6 leaves (3 from Conference and 3 from William), alternately arranged. The leaves were collected from different plants and immediately employed in the experiments. The insects were counted on each leaf every 5 minutes for 1 hour. Data were analyzed comparing the average insect number found on each cultivar by Student’s t test. These experiments were performed in September-October 2008. Choice test on leaf disks in laboratory. Each group of 6 individuals (male and female S. pyri IIIrd generation adults, raised on Abate) was placed inside a 15-cm Petri dish containing 3 leaf disks (2 cm diameter) per cultivar, alternately arranged. A total of 46 replicates were performed, for a total of 552 insects. The average number of insects found over the cultivar disks in each 5-minute interval was recorded and data were compared by Student’s t test, as previously described. Choice test on leaf disks in laboratory - insects raised on Conference. The experiments were performed as previously described, but employing male and female S. pyri IIIrd generation adults, raised on Conference cultivar (instead of Abate). A total of 46 replicates were performed, for a total of 552 insects. The average number of insects found over the cultivar disks in each 5-minute interval was recorded and data were compared by Student’s t test, as previously described. Choice test in field. A total of 15 plants for each cultivar (Conference and William) were employed, alternatively placing the grafted plants on two parallel rows, 40 cm apart. To isolate plants from attacks of other pests and from S. pyri parasites and predators, each row was covered by the anti-aphid net. The artificial infestation was performed with first generation S. pyri adults raised on Abate on 13 June 2007. Two adult insects were introduced for each plant and the infestation resulted uniform (relevant condition for experimental success). The observations started 3 days later and were performed at 3-day intervals by counting the number of adult insects on each plant and, later, also the number of nymphs derived from eggs laid by the adults initially introduced. At the end of tests the average number of insects found on each cultivar and sampling date was compared. The experiments ended on 25 July 2007 because of sharp plant decline due to heavy infestation by lace bugs. Statistical analyses were performed by Student’s t test as previously described. No-choice test in field camp. As in the previous experiments, a total of 15 one-year grafted plants of both cultivars Conference e William, arranged in parallel rows at a distance of 40 cm were employed. To isolate plants from attacks of other pests and from S. pyri parasites and predators, each row was covered by the anti-aphid net, that also prevented possible rain and wind damages. The artificial infestation was performed on 28 April 2008, employing first generation S. pyri adults raised on Abate: the infestation resulted uniform. RESULTS Feeding habits and stylet position This study performed on Stephanitis pyri allowed to record for the first time the feeding behaviour of the insect through EPG-DC recording. The data analysis allowed to characterize the EPG-DC waveform and identify the different phases. Two types of waveform were detected, waveform 1 and waveform 2, representing two different phases of insect feeding (respectively phase 2 and phase 3). Two other phases were identified: the initial one (phase 1), related to stylet penetration into leaf tissues, and the final one (phase 4), during which it is possible to observe at the stereomicroscope the stylet retrieval from the leaf tissues. Thus the EPG-DC of S. pyri is composed of four phases, two of which (phase 2 and phase 3), respectively identifiable by typical waveforms, waveform 1 and waveform 2. The total number of screenings performed and validated (including the data of comparison of phase duration between the two cultivars) was 85, each of them for a length of 8 hours. Correlation of stylet position with EPG-DC wave 1 and 2. The stylectomy in S. pyri to identify the stylet position in relation to the waveform type allowed to obtain in four experiments data on their position in relation to wave 2 (or feeding phase 3). In two cases the stylets penetrate until the borders of vascular bundles, and in the other two they penetrate the mesophyll, but do not go beyond the spongy parenchyma. These observations strongly support the hypothesis that during waveform 2 does S. pyri not perform a specialized tissue choice. Comparison of phases between Conference and William A total of 65 repeats (each lasting 8 hours) was performed using 31 adult f S. pyri or the Conference cultivar and 34 for the William one. The results of this series of experiments shows a significant difference (p = 0.014) in total duration of waveform 2, longer in William (average duration of 1906.65 seconds) in comparison to Conference (average duration 1275.73 seconds; standard error respectively 82.90 in William and 241.66 in Conference). Influence of Conference and William cultivars on S. pyri behaviour in pre-feeding phase and insect permanence on host plant Choice test on leaves in laboratory. S. pyri expresses a significant host preference for Conference leaves, already detectable after 10 minutes (an average of 1.32 individuals on Conference and of 0.8 individuals per leaf on William, p = 0.03). This difference increases with time. Choice test on leaf disks in laboratory. Again, a significant preference is expressed by S. pyri towards Conference from minute 35 to minute 40. The differences become non-significant from minute 41 to minute 55, then again are significant until the end of the experiment. Choice test on leaf disks – insects raised on Conference. The same series of experiments was repeated with insects raised on Conference cultivar. The results were very similar to those obtained in the previous experiments: Conference is the preferred host plant, with values similar to the previous ones. Choice test in field. The field experiments also support the preference of S. pyri towards Conference cultivar, with a significant difference (an average peak presence of about 65 individuals vs. about 35 in William). These results are also supported by the average number of nymphs on each plant, with a peak of almost 120 individuals per Conference plant, in comparison to about 40 individuals per William plant. No-Choice test in field. Towards the end of June a significant difference was noticed in the average number of individuals, in favour of William, but in July the significant difference favours Conference. The results also show that the peak of number of adults is reached after about five weeks on the William cultivar (average about 15 individuals per plant). The peak of the number of adult S. pyri on Conference appears later but reaches an average of almost 20 individuals per plant. Concerning the average number of nymphs per plant, it is always higher in William in comparison to Conference (respectively an average of 90 vs. 50). The average number rebalances in the last part of the experiment, reaching an average maximum peak of 120 individuals for both varieties. CONCLUSIONS - For the first time the EPG-DC waveform was recorded and characterized in all its phases in the pear leace bug, Stephanytis pyri (Fabricius) (Heteroptera, Tingidae). - Two economically relevant pear cultivars (“Conference” and “William”), common in orchards of Emilia-Romagna (Northern Italy) were analyzed for feeding habits and host choice in S. pyri. Significant difference were detected between the two cultivars in waveform 2 (feeding phase 3). - The stylet position was analyzed by optical and electron microscopy and was also determined in association to waveform 2 (feeding phase 3) - The choice tests on leaves, leaf disks and plants of both cultivars showed significant differences in host preference by. S. pyri. These difference are very likely related to morphological factors, the most relevant of which appears the leaf shape. - The field choice tests showed a host preference of to S. pyri towards Conference, when the cultivars are tested simultaneously. However, when plants are separately tested, the insects favour William for population development. The data collected in this study appear very interesting and require further investigations, especially concerning the complex relationships between S. pyri and the pear cultivar tested: these relationships may allow to identify interesting traits for agronomical practices with low environmental impact.