A lace bug as biological control agent of yellow starthistle, Centaurea solstitialis L. (Asteraceae): an unusual choice A. Paolini, 1 C. Tronci, 1 F. Lecce, 1 R. Hayat, 2 F. Di Cristina, 1 M. Cristofaro 3 and L. Smith 4 Summary The lace bug Tingis grisea Germ. (Hemiptera: Tingidae) is a univoltine sap-feeder associated with the genus Centaurea L. and distributed throughout Central and Southern Europe and the Middle East. In 2002, one Turkish population of T. grisea was selected as a potential biological control agent for yellow starthistle, Centaurea solstitialis L., (Asteraceae: Cardueae), a weed of primary concern in the USA. Field observations showed that significant damage was caused to the host plant especially when many individuals were feeding on the same plant. Life-cycle and biology observations were made to assess the duration of the five nymphal instars of T. grisea under laboratory conditions, as well as female fecundity and longevity. Starvation and oviposition no-choice tests were carried out in order to determine the host specificity of the insect. Results showed a clear oligophagous behaviour closely restricted to the genus Centaurea. In addition, among the three Centaurea spp. on which full larval development was ascertained (C. solstitialis, Centaurea sulphurea, Centaurea cyanus), yellow starthistle was clearly most suitable regarding number of eggs laid and number of adults obtained. Keywords: YST, host range, Tingis grisea. Introduction Yellow starthistle, Centaurea solstitialis L., (Asteraceae: Cardueae) is an important invasive alien weed of rangeland in the western United States and is the target of a USDA classical biological control program (Turner et al., 1995; Sheley et al., 1999; Smith, 2004; Di Tomaso et al., 2006). Six insect agents that attack C. solstitialis flowerheads have already been introduced (Cristofaro et al., 2002; Pitcairn et al., 2004), but they do not appear to be reducing the weed population sufficiently (Pitcairn et al., 2000, 2006). Therefore, it is desirable to find new agents that attack other organs of the plant or earlier phenological stages. A rust, Puccinia jaceae 1 Biotechnology and Biological Control Agency, Via del Bosco 10, 00060 Sacrofano, Rome, Italy. 2 Atatürk University, Faculty of Agriculture, Plant Protection Department, 25240 TR Erzurum, Turkey. 3 ENEA C.R. Casaccia BIOTEC, Via Anguillarese 301, 00123 S. Maria di Galeria, Rome, Italy. 4 USDA-ARS, 800 Buchanan Street, Albany, CA 94710, USA. Corresponding author: A. Paolini <a.paolini@bbca.it>. CAB International 2008 var. solstitialis Otth., has been released (Woods, 2004; Fisher et al., 2006), and two beetles are being evaluated (Cristofaro et al., 2004; Smith, 2004; Smith, 2007); all of these attack immature plants. However, it would also be useful to have an agent that stresses the plant later in the growing season, during the critical period when it is flowering and producing seed (Smith, 2004). During foreign exploration for new biological control agents in eastern Turkey in 2002, we discovered a large population of the lace bug Tingis grisea Germ. 1835 (Hemiptera: Tingidae) feeding on mature C. solstitialis rosettes and on bolting plants. In the literature, this lace bug has been reported from 11 species of Centaurea, including C. solstitialis, as well as from Crupina vulgaris L. (Stusak, 1959), a very closely related plant species (Susanna et al., 1995). Its geographical distribution is very wide: it occurs from central Spain across southern Europe to southern Russia; generally overlapping the distribution of C. solstitialis and some close relatives (Komarov, 1934; Klokov et al., 1963; Wagenitz, 1975; Dostál, 1976). Little is known about its biology (Stusak, 1959; Péricart, 1983). T. grisea is reported as univoltine and overwinters as adult. Oviposition begins in May, and 189
XII International Symposium on Biological Control of Weeds eggs are inserted into the young tissue of stems and axils. Eggs are very small; the operculum measures 0.16 0.05 mm (Stusak, 1957). In Ukraine, nymphs appear at the beginning of June and adults at the beginning of July. Five nymphal instars have been described. This paper reports results of studies on the life cycle, rearing and host-plant specificity of this insect to determine whether it warrants further evaluation as a candidate for biological control of yellow starthistle. Methods and materials Collection of insects Insects emerging from winter diapause were collected in late March of 2004, 2005 and 2006 in the vicinity of Horasan (Erzurum Region, 1600 m ASL), Eastern Turkey. In the laboratory, lace bugs were kept in a 3-l glass beaker at low temperature (8 C) and 12:12 h L/D photoperiod. Insects were allowed to feed on freshly cut leaves of C. solstitialis (US biotype) held in water vials; crumpled tissue paper was also provided as shelter for insects to rest. Laboratory rearing Insect rearing was carried out on natural substrate, using potted plants of C. solstitialis (US biotype) at 23 C to 26 C and 16:8 h L/D. Single pairs of T. grisea were confined to a portion of yellow starthistle stem anchored in a foam disk on the bottom of a 17 5 cm transparent plastic cylinder, capped with fine nylon mesh. A hole in the side, closed by a foam plug, was used for insect manipulation. After 7 days, insects were removed and transferred to another stem under the same conditions. Beginning 10 days after insect exposure, stems were cut off and daily examined under a stereo microscope to search for neonate nymphs. The same procedure was repeated several times for each pair of insects. Emerged nymphs were used for host-range, larval transfer experiments and life-cycle observations. Host specificity The host specificity of the insect was assessed by means of no-choice tests on plant species related to yellow starthistle, including US native and US commercial crops. No-choice oviposition experiment In 2004, stems of C. solstitialis (US biotype) and other test plants were exposed to a pair of insects in transparent cylinders at 23 C to 26 C and 16:8 h L/D. After 3 to 4 days, insects were removed and stems observed under the stereo microscope in order to count eggs. Because of the extremely small size of egg opercula and the complexity of stem features (tissue foldings and richness of trichomes), it was impossible to count eggs. Thus, we indirectly evaluated oviposition success by retrieving emerged nymphs from stems exposed to insects. In 2006, insects were tested on leaves instead of stems. In fact, preliminary trials carried out at the beginning of the season showed that T. grisea is able to lay eggs both on leaves and on stems. In this way, we were able to see eggs by observing leaves under the stereo microscope with backlighting. Leaves with eggs were stored in a plastic box on tissue paper until nymph emergence. Each pair was kept on yellow starthistle before and after being tested on any other plant species in order to give the insects the possibility to feed on the host plant and to be sure that females were actually ovipositing. Replicates on test plants were considered invalid if the continuity of a female s oviposition ability could not be demonstrated on C. solstitialis after a replicate with zero eggs on any non-host plant. Two to 11 specimens for each plant species were tested in 2004 and 2006; the are listed in Table 1. Larval transfer experiment Nymphs of T. grisea (two to six per replicate) were transferred to intact leaves of potted C. solstitialis and other test-plant species, confined in transparent plastic cylinders at 23 C to 26 C and 16:8 h L/D. The first observation occurred after 7 days in order to assess nymphal development and mortality. Afterwards, observations were carried out every 3 to 4 days until all the nymphs either reached the adulthood or died. For each plant species, we tested an average of five specimens. Plants tested from 2004 to 2006 are listed in Table 2. Results and discussion Life cycle Life-cycle observations, carried out during laboratory rearing and oviposition and larval transfer trials, show that under laboratory conditions (23 C to 26 C, 16:8 h L/D), first-instar nymphs emerged 10 to 12 days after oviposition. The duration of the first and second larval stages was approximately 3 to 4 days, while the development of each stage, from third to fifth instars, took 7 to 8 days. Total development was approximately 31 days (Figure 1). No-choice oviposition experiment Results of no-choice oviposition tests, performed in 2004 and in 2006, clearly showed that T. grisea oviposits most on C. solstitialis and, with limited success, on closely-related species (Table 1), including Centaurea stoebe, Centaurea cyanus, Centaurea diffusa, Centaurea sulphurea and Acroptilon repens. In 2004, oviposition occurred on only three of the eight plant species tested. The number of larvae that emerged per replicate 190
Table 1. Summary of no-choice oviposition tests carried out in 2004 and 2006. 2004 No-choice oviposition 2006 No-choice oviposition emerged larvae Valid Non-valid larvae/valid eggs laid emerged larvae % emerged larvae Valid Non-valid eggs/valid Subtribe: Centaureinae Acroptilon repens 3 5 0 0 3 0 1.7 0 Carthamus tinctorius - 5 0 3 3 0 6 0 0 0 2 5 0 0 Linoleic Carthamus tinctorius - Oleic 5 0 3 3 0 5 0 0 0 2 3 0 0 Crupina vulgaris 3 0 2 1 0 Genus: Centaurea Centaurea cyanus 3 4 3 1 1.3 5 4 1 25 2 4 2 0.5 Centaurea diffusa 3 6 1 17 3 0 2 0.3 Centaurea stoebe 4 5 2 2 2.5 Centaurea melitensis 3 0 0 0 2 1 0 0 Centaurea solstitialis 16 22 25 19 0.9 19 266 106 40 59 19 4.5 1.8 Centaurea sulphurea 2 3 0 0 2 0 1.5 0 Subtribe: Carduinae Carduus pycnocephalus larvae/valid 2 0 0 0 0 2 0 0 Cirsium brevistylum 2 0 0 0 0 2 0 0 Cirsium hydrophilum 2 0 0 0 0 2 0 0 Cirsium loncholepis 3 0 0 0 0 3 0 0 Cirsium occidentale 2 0 0 0 0 2 0 0 Cynara scolymus 4 0 2 2 0 2 0 0 0 0 4 0 0 Tribe: Heliantheae Helianthus annuus 3 0 0 0 0 3 0 0
XII International Symposium on Biological Control of Weeds Table 2. Summary of the larval transfer results conducted during 2004 to 2006. Stage of larvae transferred a Total plants tested L1 or L2 L3 or L4 Total larvae transferred a larvae transferred % developed L3 % developed adult larvae transferred % developed adult Subtribe: Centaureinae Acroptilon repens Carthamus tinctorius - Linoleic 5 30 0 0 7 29 0 12 59 Carthamus tinctorius - Oleic 6 33 6 0 6 31 3 12 64 Crupina vulgaris 3 18 0 0 1 10 0 4 28 Genus: Centaurea Centaurea americana 2 20 0 0 2 20 Centaurea cyanus 5 27 56 41 7 28 29 12 55 Centaurea diffusa 2 10 30 0 2 10 Centaurea stoebe 2 10 30 0 2 10 0 4 20 Centaurea melitensis Centaurea rothrockii Centaurea solstitialis-ca 6 32 81 63 2 15 73 8 47 Centaurea sulphurea 3 20 25 3 20 Subtribe: Carduinae Carduus pycnocephalus Cirsium brevistylum 1 10 0 0 1 10 Cirsium cymosum 2 10 10 0 2 10 Cirsium hydrophilum 2 10 0 0 4 16 0 6 26 Cirsium loncholepis 2 10 0 0 4 16 0 6 26 Cirsium occidentale 1 10 20 0 1 10 Cirsium vinaceum 2 10 0 0 2 10 Cynara scolymus 4 26 0 0 1 10 10 5 36 Tribe: Heliantheae Helianthus annuus 1 10 0 0 4 17 0 5 27 L1 First-instar nymphs, L2 second-instar nymphs, L3 third-instar nymphs, L4 fourth-instar nymphs. 192
A lace bug as biological control agent of yellow starthistle, Centaurea solstitialis L. (Asteraceae) Figure 1. Diagram of estimated development time of immature stages of Tingis grisea under laboratory conditions (23 + 3 C, 16:8 h L/D photoperiod). was 0.9 on C. solstitialis, 1.3 on C. cyanus and 2.5 on C. stoebe. The total number of nymphs emerged was considerably higher on yellow starthistle, although it was very low compared to the number of carried out (Table 1). Lack of knowledge about development times of the insect in this preliminary phase, in addition to a relatively low survival rate of the eggs on cut stems, was probably responsible for this low number of nymphs recorded. In 2006, females laid nearly 94% of their eggs on yellow starthistle, and 40% of them produced nymphs. The relatively low emergence rate can be attributed to low egg survival on cut leaves due to rapid withering and occurrence of mould. Although we found eggs on C. cyanus, C. diffusa, C. sulphurea and A. repens, nymphs emerged only from eggs laid on C. cyanus and C. diffusa. In both C. sulphurea and A. repens, leaves with eggs became mouldy several days after oviposition, thus further experiments are needed to improve the measure of nymphal emergence on these plants. Larval transfer experiment In larval development no-choice tests carried out from 2004 to 2006, nymphal survivorship was greatest on yellow starthistle (Table 2): 63% of first and second instar nymphs and 73% of third and fourth instar nymphs reached the adult stage on C. solstitialis, while a smaller percentage was able to complete the development on a small number of closely related species (C. cyanus and C. sulphurea). In general, development and survival was greater for old nymphs than for young ones, on both the target and nontarget plants. We observed development of one adult from 3rd instar on an oleic variety of Carthamus tinctorius and one on Cynara scolymus, but none of the younger (1st and 2nd instar nymphs) became adults. This insect is not a pest of either of these crops, which suggests that transfer of 1st and 2nd instars is more valid than transfer of 3rd and 4th instars. Conclusions Preliminary host-specificity results, obtained from 2004 to 2006, showed clear oligophagy by T. grisea. Among the species on which oviposition occurred (C. cyanus, C. stoebe, C. diffusa, C. sulphurea and A. repens), the target weed C. solstitialis was clearly preferred in terms of number of eggs laid and number of nymphs emerged. In addition, only a few nymphs that were transferred to non-target plants completed development on species closely related to yellow starthistle (C. cyanus and C. sulphurea). 3rd instar nymphs could sometimes develop on critical nontarget plants, such as C. tinctorius and C. scolymus, but transfer of 3rd and 4th instar nymphs represents an extreme situation that is not likely to occur in the field because the nymphs are not highly mobile. Failure of young larvae to develop on C. americana is very promising because this is the closest native North American relative to the target weed. Further feeding and oviposition trials under nochoice and choice conditions are required to better define the host range of this insect and understand if it represents a good candidate for the biological control of C. solstitialis. Moreover, laboratory tests and openfield trials are needed to evaluate its impact on the target species. Acknowledgements We are grateful to Levent Gültekin and Göksel Tozlu, Atatürk University (Erzurum, Turkey), for their support in field collections. References Cristofaro, M., Hayat, R., Gultekin, L., Tozlu, G., Zengin, H., Tronci, C., Lecce, F., Sahin, F. and Smith, L. (2002) Preliminary screening of new natural enemies of yellow 193
XII International Symposium on Biological Control of Weeds starthistle, Centaurea solstitialis L. (Asteraceae) in Eastern Anatolia. In: Proceeding of the Fifth Turkish National Congress of Biological Control. 4 7 September 2002, Erzurum, Turkey, pp. 287 295. Cristofaro, M., Dolgovskaya, M.Yu, Konstantinov, A., Lecce, F., Reznik, S.Ya., Smith, L., Tronci, C. and Volkovitsh, M.G. (2004) Psylliodes chalcomerus Illiger (Coleoptera: Chrysomelidae: Alticinae), a flea beetle candidate for biological control of yellow starthistle Centaurea solstitialis. In: Cullen, J.M., Briese, D.T., Kriticos, D.J., Lonsdale, W.M., Morin, L. and Scott, J.K. (eds) Proceedings of the XI International Symposium on Biological Control of Weeds. CSIRO Entomology, Canberra, Australia, pp. 75 80. Di Tomaso, J., Kyser, G.B. Pitcairn, M.J. (2006) Yellow starthistle Management Guide. California Invasive Plant Council. Publ. no. 2006-03. 74p. Dostál, J. (1976) Centaurea L. In: Tutin, G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M. and Webb, D.A. (eds) Flora Europaea, Vol. 4. Cambridge University Press, Cambridge, UK, pp. 254 301. Fisher, A.J., Bruckart, W.L., McMahon, M.B., Luster, D.G. and Smith, L. (2006) First report of Puccinia jaceae var. solstitialis pycnia on yellow starthistle in the United States. Plant Disease 90, 1362. Klokov, M.B., Sonsovskii, D.I., Tsvelev, N.N. and Cherepanov C.K. (1963) Centaurea. Flora URSS XXVIII, 370 579. Institutum Botanicum nomine V. Komarovii Academiae Scientiarum URSS. Editio Academiae Scientiarum Moscow, URSS. Komarov, V.L. (ed.) (1934) Flora of the U.S.S.R. Akademiya Nauk SSSR. Botanicheskii institut 28, 571 573. Péricart, J. (1983) Faune de France: France et Régions Limitrophes. Ch. 69. Hémiptères Tingidae Euro-Méditerranéens. Paris: Féderation Française des Sociétés de Sciences Naturelles. p 618. Pitcairn, M.J., Woods, D.M., Joley, D.B., Turner, C.E. and Balciunas, J.K. (2000) Population buildup and combined impact of introduced insects on yellow starthistle (Centaurea solstitialis L.) in California. In: Spencer, N.R. (ed.) Proceedings of the X International Symposium on Biological Control of Weeds. Montana State University, Bozeman, MT, pp. 747 751. Pitcairn, M.J., Piper, G.L. and Coombs, E.M. (2004) Yellow starthistle. In: Coomb, E.M., Clark, J.K., Pipe, G.L. and Confrancesco, A.F., Jr (eds) Biological Control of Invasive Plants in the United States. Oregon State University Press, Corvallis, pp. 421 435. Pitcairn, M.J., Schoenig, S., Yacoub, R. and Gendron, J. (2006) Yellow starthistle continues its spread in California. California Agriculture 60(2), 83 90. Sheley, R.L., Larson, L.L. and Jacobs, J.J. (1999) Yellow starthistle. In: Sheley, R.L. and Petroff, J.K. (eds) Biology and Management of Noxious Rangeland Weeds. Oregon State Univ. Press., Corvallis, OR, pp. 408 416. Smith, L. (2004) Prospective new agents for biological control of yellow starthistle. In: Proceedings 56th Annual California Weed Science Society. 12 14 January, 2004, California Weed Science Society, Salinas, CA, pp. 136 138. Smith, L. (2007) Physiological host range of Ceratapion basicorne, a prospective biological control agent of Centaurea solstitialis (Asteraceae). Biological Control 41, 120 133. Stusak, J.M. (1957) Beitrag zur Kenntuis der Eier der Tingiden (Heteroptera, Tingidae). Beiträge zur Entomologie 7, 20 28. Stusak J.M. (1959) Early stages of the lace bug Tingis grisea Germar (Hemiptera-Heteroptera, Tingidae). Casopis Ceskoslovenske Spolecnosti Entomologiske, Acta Societatis Entomologicae Cechosloveniae 56(2), 181 200. Susanna, A., Garcia-Jacas, N., Soltis, D.E. and Soltis, P.S. (1995) Phylogenetic relationships in tribe Cardueae (Asteraceae) based on ITS sequences. American Journal of Botany 82, 1056 1068. Turner, C.E., Johnson, J.B. and McCaffrey, J.P. (1995) Yellow starthistle. In: Nechols, J.R., Andres, L.A., Beardsley, J.W., Goeden, R.D. and Jackson, C.G. (eds) Biological Control in the Western United States: Accomplishments and benefits of regional research project W-84, 1964 1989. University of California, Division of Agriculture and Natural Resources, Oakland Publ. 3361, pp. 270 275. Wagenitz, G. (1975) 79. Centaurea L. In: Davis, P.H. (ed.) Flora of Turkey and the East Aegean Islands, Vol. 5. Edinburgh University Press, Edinburgh, pp. 465 585. Woods, D.M. (2004) Development and release of a plant pathogen as a new biological control of yellow starthistle. Cal-IPC News 12(2), 6 7. 194