Understanding the fitness, preference and performance of specialist herbivores of the Southern African biotype of Chromolaena Odorata (Asteraceae), and impacts on phytochemistry and growth rate of the plant.

Abstract

The invasiveness and negative impacts of the alien shrub, Chromolaena odorata (L.) (Asteraceae) in South Africa resulted in the initiation of a biological control programme against the weed in the late 1980s. After the release of seven biocontrol agents, only two have successfully established to date viz. a leaf mining fly, Calycomyza eupatorivora Spencer (Diptera: Agromyzidae) and a moth with defoliating larvae, Pareuchaetes insulata Walker (Lepidoptera: Erebidae: Arctiinae). Surveys conducted suggested that C. odorata densities seem to be low in KwaZulu-Natal (KZN) province where P. insulata is present whilst its infestation is increasing in other places such as Limpopo province where the moth is absent. This study aimed to examine the life history traits, preference and performance of two biocontrol agents, viz. Dichrorampha odorata (Lepidoptera: Tortricidae) and Polymorphomyia basilica (Diptera: Tephritidae). A further objective of the study was to measure the effects of one of the established agents, P. insulata, on the competitive ability and defence mechanism of C. odorata by indirectly testing the predictions of the Evolution of Increased Competitive Ability (EICA). Studies of life history traits of D. odorata in the laboratory indicated that the moth possesses good biological control attributes such as short-lived adults with high mating success, fecundity and egg hatchability. Of the 34 asteraceous plants subjected to larval no-choice tests, only C. odorata could sustain complete development of D. odorata to adulthood, although there was slight initial boring on 14 test species (plus C. odorata). Adult no-choice tests using seven test-plant species that were damaged in larval no-choice tests were consistent with the earlier trials, with larval damage, pupae and adults of D. odorata recorded from only C. odorata. This demonstrated that only C. odorata is a suitable host for D. odorata in South Africa and permission for the release of this first shoot-tip attacking agent was granted for biocontrol of C. odorata in South Africa. To predict the efficacy of D. odorata as a biological control of C. odorata, a 9-month laboratory study was carried out. Plant growth metrics were compared across three treatments i.e. 0, 50 and 100% where newly hatched D. odorata larvae were inoculated onto the shoot tips of C. odorata. At all treatment levels, the basal stem diameter of C. odorata was not affected by D. odorata larval feeding whilst the height of the main shoot and flower production of C. odorata were reduced at 50 and 100% relative to the control treatment. In general, the impacts of D. odorata on the weed were relatively small even though statistically significant, suggesting that the moth will modestly reduce the height and flower production of C. odorata. Positive biological characteristics of P. basilica include a high rate of increase, long-lived and mobile adults, the ability of females to produce viable offspring without repeated mating, the ability of adults to eclose from galls on dry stems and the production of several generations per year. Thirty-two asteraceous plants were investigated to determine host specificity of P. basilica in single-choice adult tests and using single pairs of adults in no-choice tests, under laboratory conditions. Oviposition and larval development through to adulthood occurred on three other South American and on two South African species; one in the same tribe Eupatorieae, closely related to C. odorata and another one in the Astereae, less closely related to the weed, but both at a lower and slower rate. Females tended to retain their eggs under no-choice conditions in the presence of an unsuitable host, and to compensate by ovipositing at a higher rate when presented later with a C. odorata plant. Overall, this study predicts the ability of P. basilica to stretch to areas where P. insulata has failed to establish and supports the suitability of P. basilica for release in South Africa. To determine the mechanism behind the decrease of C. odorata densities in KZN province, where the specialist herbivore P. insulata is present, compared to Limpopo province, where the weed is increasing and the moth is absent, the Evolution of Increased Competitive Ability (EICA) hypothesis was indirectly tested on plant defence and growth rate metrics. Inconsistent with EICA, total phenolics and tannins were generally higher in Thohoyandou (Limpopo province) (without P. insulata) and Komatipoort (Mpumalanga province) (with P. insulata) and lower in Pietermaritzburg (KZN province) (without P. insulata) and Umkomaas (KZN province) (with P. insulata). Flavonoids varied between the four locations, with higher concentrations in Komatipoort compared to Thohoyandou and Umkomaas, but not different to Pietermaritzburg. Growth parameters such as stem diameter, number of shoots and number of flowering shoots from the garden experiment, supported the prediction of EICA, as plants from the Thohoyandou and Pietermaritzburg sites, where P. insulata is absent, showed stronger growth and reproductive potential. This study demonstrates the possible role of P. insulata in the decrease in population of C. odorata where the moth has persisted and suggests that other biotic and abiotic factors could be responsible for the unpredicted results for phytochemistry assays. The second part of the EICA hypothesis posits that “specialist herbivores will demonstrate improved performance on plant individuals originating from an area where plants have been introduced”. Consistent with EICA, Pareuchaetes insulata immature stages (newly hatched larvae-adult eclosion) that fed on leaves from Umkomaas, had prolonged development compared to larvae that were fed on leaves from Thohoyandou and Pietermaritzburg, and Komatipoort. Larvae and pupae that fed on the leaves from shade from Komatipoort had developmental trends intermediate between larvae feeding on the leaves from the shade from Thohoyandou and Umkomaas. Overall survival was lowest on leaves of plants obtained from Komatipoort. Contrarily, location did not appear to influence pupal mass but this variable was higher in plants in the full sun. In sum, the existing reassociation time may not be enough for evolutionary changes to have occurred in C. odorata defence and P. insulata response to plant evolution, and could explain the inconsistency in some P. insulata performance parameters on infested and uninfested populations of C. odorata. The roles of pyrrolizidine alkaloids (PAs) as plant defences and in P. insulata mating behaviour are well known. The PAs in the roots of the southern African biotype (SAB) of C. odorata were therefore examined. Two PAs, rinderine and its stereoisomer N-oxide intermidine, were isolated from the roots of the SAB of C. odorata using GC-MS. The structures and configuration were confirmed by chemical and spectroscopic methods, especially one- 1H dimensional NMR analysis. Therefore, confirmation of rinderine and intermidine in C. odorata in this study substantiate the establishment and spread of P. insulata in southern Africa due to, among other factors, reduced predation through defence by sequestered PAs. This study demonstrated positive biological characteristics and high preference and performance of both the moth with shoot-tip boring larvae D. odorata and the stem-galling fly P. basilica on C. odorata compared to non-target plants, which highlights positive prospects for the biological control programme of C. odorata in South Africa. This study reports for the first time two pyrrolizidine alkaloids viz. intermidine and rinderine in southern African C. odorata. Aspects of EICA were not straightforward; however, this study showed the contribution of P. insulata to the reduction of C. odorata where the moth is present and further provides direction for future research for the biological control of C. odorata in South Africa.