BioControl (2014) 59:367–375 DOI 10.1007/s10526-014-9573-z The impact of the flower mite Aceria acroptiloni on the invasive plant Russian knapweed, Rhaponticum repens, in its native range Ghorbanali Asadi • Reza Ghorbani • Massimo Cristofaro • Philipp Chetverikov • Radmila Petanovic´ • Biljana Vidovic´ • Urs Schaffner Received: 26 October 2013 / Accepted: 13 March 2014 / Published online: 27 March 2014 Ó International Organization for Biological Control (IOBC) 2014 Abstract Rhaponticum repens (L.) Hidalgo is a clonal field site revealed that A. acroptiloni was by far the Asteraceae plant native to Asia and highly invasive in dominant mite species. We conclude that the mite A. North America. We conducted open-field experiments in acroptiloni is a promising biological control candidate Iran to assess the impact of the biological control inflicting significant impact on the above-ground biomass candidate, Aceria acroptiloni Shevchenko & Kovalev and reproductive output of the invasive plant R. repens. (Acari, Eriophyidae), on the target weed. Using three different experimental approaches, we found that mite Keywords Above-ground biomass Á Acari Á attack reduced the biomass of R. repens shoots by Acroptilon Á Asteraceae Á Classical biological 40–75 %. Except for the initial year of artificial infesta- control Á Pre-release studies Á Seed production tion by A. acroptiloni of R. repens shoots, the number of seed heads was reduced by 60–80 % and the number of seeds by 95–98 %. Morphological investigations of the Introduction mite complex attacking R. repens at the experimental The aim of pre-release studies in classical biological weed control projects is not only to experimentally Handling editor: John Scott. G. Asadi Á R. Ghorbani P. Chetverikov Department of Agronomy, Faculty of Agriculture, Department of Invertebrate Zoology, Saint-Petersburg Ferdowsi University of Mashhad, P.O. Box 91775-1163, State University, Universitetskaya nab., 7/9, Mashhad, Iran St. Petersburg 199034, Russia e-mail: [email protected] R. Petanovic´ Á B. Vidovic´ R. Ghorbani Department of Entomology and Agricultural Zoology, e-mail: [email protected] Faculty of Agriculture, University of Belgrade, Nemanjina 6, 11080 Belgrade-Zemun, Serbia M. Cristofaro e-mail: [email protected] ENEA Casaccia, UTAGRI-ECO, via Anguillarese 301, 00123 Rome, Italy B. Vidovic´ e-mail: [email protected] e-mail: [email protected] P. Chetverikov U. Schaffner (&) Zoological Institute, Russian Academy of Sciences, CABI, Rue de Grillons 1, Dele´mont, Switzerland Universitetskaya nab. 1, St. Petersburg 199034, Russia e-mail: [email protected] e-mail: [email protected]; [email protected] 123 368 G. Asadi et al. assess the host-range of biological control candidates, successful in the field (Coombs et al. 2004). In but also to identify those specialist herbivores that 2010, two additional biological control agents were impose significant damage upon the target weed and released in the USA and Canada, i.e. the gall wasp thereby have the potential to reduce weed densities in Aulacidea acroptilonica V.Bel. (Hymenoptera, Cyn- the invaded range (Pearson and Callaway 2003). ipidae) and the gall midge Jaapiella ivannikovi Predicting the efficacy of biological control candidates Fedotova (Diptera, Cecidomyiidae). Both agents can is far from an easy task (Mu¨ller-Scha¨rer and Schaffner significantly reduce the reproductive output of R. 2008), but observational and manipulative field repens (Djamankulova et al. 2008). While the two experiments in the native range, paired with informa- species have established in North America, their tion on the population dynamics of the weed in the population densities have not yet reached a level invaded range, can provide some estimates on the where they significantly impact R. repens at local or candidate’s potential impact on the invasive weed. even regional scale (L. Baker, personal Comparing quantitative data on the impact of the communication). biological control agents in the native range with those Another specialist natural enemy considered for its collected post-release can also help refining and potential as a biological control agent against R. improving pre-release efficacy studies in biological repens is the eriophyid mite Aceria acroptiloni control. Shevchenko & Kovalev (Acari, Eriophyidae) (Schaff- Russian knapweed, Rhaponticum repens (L.) ner et al. 2013). Here we report the results of open- Hidalgo (Acroptilon repens (L.) DC.) is a clonal field experiments on the impact of A. acroptiloni on Asteraceae plant. The native range includes Armenia, the target weed conducted in the native range. Since Turkmenistan, Uzbekistan, Kazakhstan and parts of several other mite species are reported from R. repens Turkey, Iran, Russia, China and Mongolia (Watson (Schaffner et al. 2013), we compared the morphology 1980). It was accidentally introduced into North of the mites found on the experimental plants to clarify America in the late 19th century as a contaminant of whether any impact observed in these studies could alfalfa (Medicago sativa L.) seed (Watson 1980). To indeed be attributed to A. acroptiloni. date, R. repens has spread to 45 of the 48 contiguous states of the USA (Zimmerman and Kazmer 1999) and is a declared noxious weed in 23 US states and Materials and methods four Canadian provinces (Rice 2013). There is growing ecological evidence that many North Amer- The target weed ican plants are poorly adapted to R. repens (Ni et al. 2010). For example, the biomass of native species in Rhaponticum repens is a long-lived, rhizomatous R. repens stands was 25–30 times lower in north- perennial that propagates by vegetative and sexual western USA than in Uzbekistan (Callaway et al. means (Watson 1980). In North America, R. repens 2012). Experimental addition of native plants as attains significantly higher above-ground biomass and seeds significantly increased their abundance in R. cover than in Uzbekistan (Callaway et al. 2012). Also, repens stands in northwestern USA, but the propor- North American populations produce about four times tion of their biomass relative to total biomass more seeds than those in the native range (J. Littlefield remained over an order of magnitude lower than and U. Schaffner, unpublished results), which is at that of native Asian species in R. repens stands in least partly due to significant herbivore pressure on Uzbekistan (Callaway et al. 2012). seed output in the native range (Callaway et al. 2012). In the 1970s, first attempts to control R. repens While recruitment of seedlings within established R. by biological means led to the release of the gall- repens patches is rare, it is the most important forming nematode Subanguina picridis Kirjanova mechanism by which new sites are colonized within (Tylenchida, Anguinidae). However, although labo- the invaded range. Viable seeds are a common ratory experiments suggested that the agent can occurrence in the faeces of cattle and wildlife which have a considerable impact on growth and seed appears to play an important role in spreading output of R. repens, it has not proven to be R. repens (R. Lang, unpublished report). 123 The impact of the flower mite Aceria acroptiloni 369 The mite species Medicago sativa L., Alhagi maurorum Medik and various Bromus spp. In approach A, 20 randomly Aceria acroptiloni was originally described by Shev- selected R. repens shoots growing at the experimental chenko & Kovalev (Kovalev et al. 1974), based on site were selected for mite infestation, and another 20 specimens collected in Uzbekistan and the Ukraine. were selected as control plants and kept free of mite The host-range of this biological control candidate attack by spraying once every two weeks with the appears to be restricted to R. repens (Schaffner, acaricide Propargite (Omite Ò 57E; Shimagro-Yazd, unpublished results). According to our observations, Iran; concentration 0.001 %), a non-systemic acaricide during the vegetative period all life stages of A. with the active ingredient 2-(4-tert-butylphenoxy)cyclo- acroptiloni can be found on above-ground plant parts, hexyl prop-2-yne-1-sulfonate which is widely used for first in lateral vegetative buds and later mainly in controlling phytophagous mites in crop fields. flower/seed buds. In October, no live mites were found All infested shoots were sprayed with an equivalent in the dry seed buds, but some mites were recorded amount of water. Shoots assigned to the ‘mite attack’ below-ground on dormant root buds, indicating that A. treatment were experimentally inoculated with A. acroptiloni hibernates below-ground. acroptiloni in late June 2011 by shaking three mite- Preliminary morphological studies of mites found infested flower heads over each of the 20 shoots. on R. repens revealed that in Iran more than one mite Inspection under the microscope revealed that flower species are associated with R. repens (Chetverikov heads were usually colonized by hundreds or thousands et al. 2012). We therefore randomly collected 20 of mites (Asadi and Schaffner, unpublished results). infested seed heads from R. repens plants in a field In approach B, 20 infested and 20 non-infested margin 5 km NE of Shirvan, Iran, from where the shoots from site 5 km NE of Shirvan with a natural mites were collected for the impact experiment (see mite population were transplanted to the experimental below), and morphologically analyzed 25–65 mites site in a completely randomized design on the campus per flower head. Identification of the mites was of Shirvan University. This approach was chosen conducted at the Center of Microscopy and Micro- because in previous years the site with the natural mite analysis of Saint-Petersburg State University (Russia) population was heavily grazed by sheep and goats, and using a phase contrast light microscopy (PC LM) and it was therefore uncertain whether we would find confocal laser scanning microscopy (CLSM). CLSM enough naturally infested plants at this site in autumn was carried out using a Leica TCS SP2 spectral to compare the impact between experimentally and confocal and multiphoton system microscope at an naturally infested plants.