Computational protocol: The Only African Wild Tobacco, Nicotiana africana: Alkaloid Content and the Effect of Herbivory

Similar protocols

Protocol publication

[…] Experiments were carried out in a greenhouse at the University of Pretoria. Efforts to grow N. africana plants in the field at different localities in the Pretoria region proved fruitless.We randomly assigned plants to the following treatments: 1) control (no larvae), 2) one larva per plant, 3) three larvae per plant, and 4) six larvae per plant (). The treatments were positioned in a randomized block design and each treatment was replicated five times. Larvae of H. armigera are cannibalistic and were therefore placed singly on individual leaves starting from the second-youngest expanded leaf and on successively older leaves when more than one larva was used in the treatment. The “top” expanded leaf was always undamaged and each plant had an undamaged “middle” and “bottom” leaf, except plants that had six larvae on them since on these plants the “middle” leaf would be damaged by larval feeding (). All leaves with larvae were enclosed in gauze bags to prevent the larvae from escaping. Leaves on control plants were also enclosed in gauze bags to maintain the same experimental conditions. Larvae were allowed to feed freely on their allocated leaves for four days, after which they were removed; none of the larvae died during the four days. The relative growth rate of larvae during the four-day feeding period on N. africana plants was determined as (final mass – initial mass)/initial mass. Leaves were photographed before and after herbivory to record larval feeding, and the image analysis software ImageJ 1.45s (Wayne Rasband, National Institutes of Health, USA, http://imagej.nih.gov/ij) was used to determine the amount of leaf area consumed.Leaves were excised at the base of the petiole using a razor blade six days after larvae were removed. The time was chosen because nicotine levels in other Nicotiana species (i.e. N. attenuata, N. sylvestris) have been found to peak after five days and remain elevated for about ten days after induction , . Leaves were harvested in the morning between 08h00 and 10h00. The following leaves were harvested from each plant: all leaves damaged by herbivory, the top leaf (first unfolded leaf more than 4 cm in length), one middle leaf (approximately leaf six counting from the top) and the oldest bottom leaf, unless this had already senesced, in which case the next oldest leaf was chosen. Harvested leaves were weighed and immediately frozen in liquid nitrogen, ground to a fine pulp using a pestle and mortar, and stored at −20°C in glass vials until they could be freeze-dried before alkaloid extraction.Depending on availability, we collected nine or ten flowers from each of seven plants. Flowers were excised at the peduncle and the floral tissue thus included sepals, petals, and all of the reproductive parts. Flowers from each plant were pooled and thereafter processed in the same manner as the leaves.Nectar was collected from all flowering plants immediately before the leaves were harvested. This was done by inserting a 75 µl glass microcapillary tube near the base of the corolla; its volume was measured from the column length, and then pooled for each plant. The nectar sugar concentration was measured using a hand-held refractometer (Eclipse 45–81, Bellingham & Stanley Ltd., UK). The nectar samples were also stored at −20°C until they could be freeze-dried before alkaloid extraction. [...] Data were tested for normality and homogeneity of variances. The relative growth rate of larvae was analyzed with a generalized linear model (GLM) with a gamma distribution for errors and a log link function. Means were separated using t probabilities of pairwise differences. The effect of larval density on the leaf area consumed was analyzed by analysis of variance (ANOVA). Means were separated using Fisher’s protected least significant difference (LSD) test. Generalized linear modelling using a log link function with the gamma distribution was used to test for differences in nicotine, nornicotine and anabasine concentrations in top, middle and bottom leaves, as well as the effect of different herbivore densities on the concentration of each alkaloid in a) damaged and undamaged middle leaves and b) undamaged top and undamaged bottom leaves of herbivore damaged compared to undamaged plants. Means were separated using t probabilities of pairwise differences. The significance level was set at P<0.05 for all analyses. Data were analyzed with GenStat and SPSS (IBM SPSS Statistics, version 21).The sole alkaloid recorded in flowers (four out of seven samples) was nornicotine, and this at very low concentrations only. The nectar of N. africana contained no alkaloids at all. Therefore no statistical analyses were performed on the flower and nectar samples. […]

Pipeline specifications

Software tools ImageJ, SPSS
Applications Miscellaneous, Microscopic phenotype analysis
Organisms Nicotiana tabacum, Helicoverpa zea, Helicoverpa armigera
Chemicals Anabasine, Nicotine