Host-Specificity Testing of Prospodium Transformans
Total Page:16
File Type:pdf, Size:1020Kb
Host-specificity testing of Prospodium transformans (Uredinales: Uropyxidaceae), a biological control agent for use against Tecoma stans var. stans (Bignoniaceae) A.R. Wood1 Summary Yellow bells, Tecoma stans (L.) Juss. ex Humb., Bonpl. & Kunth (Bignoniaceae), originally from Meso- and South America, is an emerging weed in the warm, moist regions of South Africa. The mi- crocyclic, gall-inducing, rust fungus Prospodium transformans (Ellis & Ever.) Cummins (Uredinales: Uropyxidaceae) is a Neotropical pathogen of T. stans. It has been observed to be damaging to its host under natural conditions and is being considered as a biological control agent for use against this weed in South Africa. Two isolates were collected, one from Guatemala and the other from southern Mexico, and established and maintained in quarantine in South Africa. Host-specificity testing was conducted using both isolates against 14 species of Bignoniaceae (eight indigenous to southern Af- rica) and a further eight indigenous species in closely related families. No symptoms were produced on any of these species, except for small chlorotic spots on Fernandoa magnifica Seem. Sporulating galls were produced on all control T. stans plant leaves, petioles and stems. Leaves of the Bigno- niaceae plants tested were examined microscopically at 7 days after inoculation. Fungal colonies of approximately 300 μm in diameter developed in control T. stans leaves, but the rust did not colonize any other species. Therefore, P. transformans is considered safe for introduction into South Africa, and permission for its release will be sought. Keywords: South Africa, emerging environmental weed, yellow bells. Introduction variable, intergrading in places with the other two de- scribed varieties (var. velutina DC. and var. angustata Tecoma Juss. (Lamiales, Bignoniaceae) is a genus of Rehder; Gentry, 1992). 14 species, mainly occurring in the Neotropics but with This plant is naturalized in South Africa, where it two species in Africa (Gentry, 1992). The genus can invades natural and disturbed vegetation and is there- be divided into two groups, one with narrow tubular fore a declared weed (Henderson, 2001). Although not bird-pollinated flowers and the other with capanulate yet regarded as a major weed, it is considered to have bee-pollinated flowers. The two African species are in- the potential to invade a large proportion of the country cluded in the former group, which is the more diverse. (Nel et. al., 2004). It is currently increasing in abun- The latter group includes Tecoma stans (L.) Juss. ex dance and has been chosen as a target of a biological Humb., Bonpl. & Kunth and three other more narrowly control programme aimed at preventing it from emerg- distributed segregate species (Gentry, 1992). ing as a weed of national importance (Olckers, 2004). T. stans var. stans is a small tree with a widespread Prospodium Arth. (Uredinales: Uropyxidaceae) is a natural distribution, occurring throughout Mesoamerica Neotropical genus of about 50 species predominantly and the Caribbean, as well as much of South America parasitizing members of the Bignoniaceae, with the (Gentry, 1992). Within this range, it is morphologically rest on the Verbenaceae (Cummins and Hiratsuka, 2003). One species, Prospodium tuberculatum (Speg.) Arth., has been introduced into Australia for the bio- logical control of Lantana camara L. (Tomley and Rid- 1 ARC-Plant Protection Research Institute, P. Bag X5017, Stellenbosch, 7599, South Africa <[email protected]>. ing, 2002), and another, P. tumefaciens Lind., has been © CAB International 2008 proposed as a potential agent for use against Aloysia 345 XII International Symposium on Biological Control of Weeds gratissima (Gill. et Hook.) Troncose in the USA (Cordo Lamiales (Table 1) were inoculated in the same manner and DeLoach, 1995). as above. Six plants of each species in the Bignoniaceae Three Prospodium spp. have been recorded as occur- were inoculated, three using the southern Mexican and ring on T. stans, namely the macrocyclic P. appendicu- three using the Guatemalan isolates. Three plants of latum (Wint.) Arth. and the microcyclic Prospodium each species in the other families were inoculated with transformans (Ellis & Ever.) Cummins and P. elegans only the isolate from southern Mexico. The plants were (Schroet.) Cummins. The latter two are presumed to then observed for gall development and sporulation for have been derived by a contraction of the life cycle of 1 month after inoculation. For every batch of plants P. appendiculatum (Cummins, 1940). All three may be inoculated, a plant of local T. stans was inoculated in considered as potential biological control agents for the same manner, at the same time. Only if sporulating use against T. stans in South Africa. One of these, P. ap- galls developed on these control plants were the results pendiculatum, is adventitious in Brazil and is currently recorded, otherwise the plants were re-inoculated. Ev- being assessed for its effectiveness as a biological con- ery plant was inoculated twice, the second time on new trol agent against T. stans in that country (Vitorino et. growth not previously inoculated. al., 2004). In its native range (Caribbean basin, Guatemala, Microscopic examination Mexico), P. transformans has only been recorded from T. stans var. stans and var. velutina (as T. mollis Humb., Two plants of each of the tested Bignoniaceae spe- Bonpl. & Kunth) (Cummins, 1940). This rust fungus cies were inoculated as above. Seven days after inocu- causes galls up to 3 cm in diameter on petioles, stems lation, two leaves from each plant were harvested and and seed pods, on which initially pycnia then telia de- prepared for microscopic examination using the whole- velop. These are the only two stages of this species’ life leaf clearing and staining technique of Bruzzese and cycle. The teliospores may germinate as soon as they Hassan (1983). The stained leaves were examined at develop (Shuttleworth, 1953). Because of its known, 400´ magnification for penetration and development of limited host range and the damage that it causes to its mycelium by P. transformans. A control plant of local host plant in its native range, studies were undertaken T. stans was included for each inoculation. to assess the suitability of introducing P. transformans in South Africa for the biological control of T. stans Table 1. List of plant species included in host-specificity var. stans. Results reported in this paper deal with host- testing of Prospodium transformans. specificity testing of P. transformans before seeking approval for release of this species in South Africa. Plant species Origin Bignoniaceae Fernandoa magnifica Seem. Af Methods and materials Fernandoa sp. Af Jacaranda mimosifolia D. Don. e Source and maintenance of cultures Kigelia africana (Lam.) Benth SA Macfadyena unguis-cati (L.) A.H. Gentry e Two isolates of P. transformans, originally collected Markhamia obtusifolia (Baker) Sprague Af from southern Mexico and Guatemala, were main- M. zanzibarica (Bojer ex DC.) K. Schum. SA tained in the quarantine glasshouses at ARC-Plant Pro- Podranea ricasoliana (Tanfani) Sprague SA tection Research Institute, Stellenbosch, South Africa, Pyrostegia venusta (Ker Gawl.) Miers e by repeated inoculation of potted T. stans plants. The Rhigozum obovatum Burch. SA plants were grown from seed collected in South Africa. Spathodea campanulata P. Beauv. Af These two isolates were kept in separate glasshouses Tecoma capensis (Thunb.) Lindl. SA to prevent cross-contamination. Plants were inoculated T. stans (L.) Juss. ex Humb., Bonpl. & Kunth var. stans e by dusting dry teliospores on the petioles and adaxial Acanthaceae surfaces of immature leaflets (pinnae) using a small Duvernoia adhatodoides E. mey. ex Nees SA paintbrush, spraying the leaflets with water using an air Mackaya bella Harv. SA brush or atomizer until very small droplets were vis- Oleaceae ible to the naked eye, and then sealing the plants within Jasminum multipartitum Hochst. SA a plastic bag. The inoculated plants were placed in an Schrophulariaceae incubator at 18°C (Shuttleworth, 1953) for 48 h and Freylinia tropica S. Moore SA then transferred to the quarantine glasshouses with a Halleria lucida L. SA day/night temperature cycle of 25/19°C. Verbenaceae Lantana rugosa Thunb. SA Lippia rehmania H. Pearson SA Host-specificity testing Lippia scaberima Sond. SA Plants of indigenous and locally cultivated species SA Indigenous to South Africa, Af indigenous elsewhere in Africa, e in the Bignoniaceae and other selected families of the exotic to Africa. 346 Host-specificity testing ofProspodium transformans Results naturalized in South Africa) and indigenous represen- tatives of other families in the Lamiales all showed Host-specificity testing no symptoms of infection. The only plant tested that showed any symptoms (chlorotic spots) was F. mag- On the control T. stans plants, chlorotic flecks were nifica. Microscopic examination revealed no fungal visible on the leaf blades approximately 5 days after mycelium associated with these chlorotic spots; rather, inoculation. Galls began to develop on these flecks dense parenchyma and accumulated crystals occurred. soon after; small galls on the leaf blades (reaching ap- These probably indicate that the chlorotic flecks were proximately 5 mm in diameter after a month) but larger due to a plant defence reaction. galls on the petioles or stems (up to 30 mm long). Pyc- Because of the high level of host specificity demon-