The aim of this study was to investigate the role of the amino acid permease gene AAP6 in regulating phloem amino acid composition and then to determine the effects of this altered diet on aphid performance. A genotype of Arabidopsis thaliana (L.) was produced in which the function of the amino acid permease gene AAP6 (At5g49630) was abolished. Plants homozygous for the insertionally inactivated AAP6 gene had a significantly larger mean rosette width than the wild type and a greater number of cauline leaves. Seeds from the aap6 mutant were also significantly larger than those from the wild-type plants. Sieve element (SE) sap was collected by aphid stylectomy and the amino acids derivatized, separated, and quantified using Capillary Electrophoresis with Laser Induced Fluorescence (CE-LIF). In spite of the large variation across samples, the total amino acid concentration of SE sap of the aap6 mutant plants was significantly lower than that of the wild-type plants. The concentrations of lysine, phenylalanine, leucine, and aspartic acid were all significantly lower in concentration in the aap6 mutant plants compared with wild-type plants. This is the first direct demonstration of a physiological role for an amino acid transporter in regulating SE composition in vivo. The amino acid availability in sieve element sap is thought to be the major limiting factor for aphid growth and reproduction. Despite the changes in their diet, the aphid Myzus persicae (Sulzer) displayed only small changes in feeding behaviour on mutant plants when measured using the Electronic Penetration Graph (EPG) technique. Salivation by the aphid into the SE (E1 phase) was increased on mutant plants but there was no significant effect on other feeding EPG behaviours, or in the rate of honeydew production. Consistent with the small effect on aphid feeding behaviour, there was only a small effect of reduced sieve element amino acid concentration on aphid reproduction. The data are discussed in relation to the regulation of phloem composition and the role of phloem amino acids in regulating aphid performance.

The effects of drought and elevated CO 2 on the performance of sap-feeding aphids is considered. It is assumed that, for these stressors, the major influence will act through altering host plant composition and therefore diet. Changes in the plant may affect the ability to locate phloem tissues, while changes in composition of the sieve element forming the aphid diet may have more direct effects. It is concluded that plant response to conditions where carbon is present in excess (elevated CO 2 ) or its consumption is exceeded by its availability (drought) is heterogeneous at the cellular level. The complexities of the response of the plant to components of climate change are paralleled by the diversity of the responses of aphids to drought and elevated CO 2 . Potential control points are discussed and it is concluded that current knowledge, both descriptive and mechanistic, supports the view that it is unreasonable to expect that a single plant component can predict the general response of aphids to climate change. Instead, it is more likely that aphids use a variety of cues when interacting with their host plants, and individual species respond to changes in their diet differently. Further work examining the response of both plant and aphid transcriptome and metabolome will support or contradict this hypothesis.

We have developed an Arabidopsis thaliana / Myzus persicae model system to allow the dissection of plant/insect interactions at a molecular genetic level. This allows the examination of the role of single plant genes in the interaction between the plant and an aphid. Our initial studies have exploited an Arabidopsis genotype in which the function of the amino acid transporter ANT1 has been abolished. This mutation results in a change in the proportions of several amino acids within the phloem sieve elements (SEs) resulting in an increase in the proportion of essential amino acids. This has been measured using aphid stylectomy to collect SE samples, followed by a novel micellar electrokinetic chromatography method for amino acid analysis. The SE content represents the aphid's diet, and use of electrical penetration graph technology and honeydew clocks have demonstrated that this altered diet results in a change in the feeding rate of the aphid. Balance sheets can be produced to show the amount (nmoles/24 h) of each of 18 amino acids taken up and excreted by aphids feeding on wild type and ant1 mutant plants. The data show that aphids feeding on the ant1 mutant take up larger amounts of amino acids. However, we could not detect any effect on the reproductive rate of the aphids. The results show that, under experimental conditions, this model system can be used to identify plant genes that control the behaviour and fecundity of an insect pest.

Amino acids were derivatised with 4-fluoro-7-nitrobenzo-2,1,3-oxadiazol (NBD-F), separated by micellar electrokinetic chromatography (MEKC), and detected by argon-ion (488 nm) laser-induced fluorescence. The optimised MEKC background electrolyte conditions were: 40 mM sodium cholate, 5 mM beta-cyclodextrin in 20 mM aqueous borate buffer, pH 9.1, with 7% v/v acetonitrile. Using these conditions, 19 amino acids were separated within 17 min. The limits of detection were in the range of 7.6-42.2 pmol/mL and limits of quantitation from 0.05-0.14 nmol/mL. The method was systematically validated for injection volume error, migration time variation, calibration linearity, accuracy, precision, and recovery. Nanolitre volume samples of phloem sap of individual sieve element cells from the plant Arabidopsis thaliana and honeydew from the aphid Myzus persicae were directly analysed with this method. Quantitative amino acid concentrations in these two biological matrices were profiled for the first time. This method is particularly important because it allows the complete profile of the amino acids obtained from individual phloem elements, allowing cell to cell and plant to plant variation to be quantified, which to date has not been possible with Arabidopsis thaliana.