Microsatellite marker analysis
The multilocus microsatellite analysis, using six microsatellite loci, namely myz9, M40, M35, M49, M63, and M86, defined 21 clones amongst the M. persicae samples collected in Scotland in 1995 and 2003–2005. Fourteen clones comprised over 98% of the collections. An additional seven clones were collected in a single year, usually only once, and are the subject of a separate study. In accordance with the convention adopted by Fenton et al. (2005), the 14 clones were designated A–N (Table 1).
Table 1. The clones found in Scotland from 1995 to 2005
Intraclonal variation was identified at loci M49 and M86. In clone L, the M49 locus (ACn) always contained an allele of 138 bp, and another allele of either 156 or 160 bp, with the latter being present in 75% of individuals (results not shown). The L clonal lineages were co-localized and were identical for other markers, which is strong evidence that they share a common asexual ancestor. The 156 bp allele was found in other Scottish clones, but the 160 bp allele was not. The 156 bp allele therefore appears to be shared with other clones colonizing Scotland. This would be consistent with the original asexual lineage of clone L having a 156 bp allele and the 160 bp allele mutating from this to form the now more common clone L lineage.
In clone I, there was consistent variation in one allele at both M49, either 203 or 207 bp with a 153 bp allele always present, and M86 (CAn), 125 or 127 bp with a 140 bp allele always present. Unlike the variation in clone L, the 127 and 207 bp variants of clone I were rare, being found in only three samples. In one case, there were a number of individuals in the sample and these were all characterized and found to carry both mutations. This confirmed that the size variants were present and in the same clonal lineage and that this is a double mutation. These lineages had identical microsatellite patterns for the other loci examined, which is strong evidence that they share a common asexual ancestor. As with lineages in clone L, one of the variants at M63 (125) was shared with other Scottish clones but the other rare one (127) was not. This implies a direction of mutation with a common allele changing to a rare allele in one of the asexual lineages. Both variants at locus M49 in the lineages of I were not shared with any other Scottish clone, although it can still be suggested that a common allele (203) mutated to a rare allele (207) in an asexual lineage.
Spatial and seasonal distributions of clones from field samples
Clones C, I and J were present in each year and area sampled, apart from clone J which was absent from Fife in 2004 (Fig. 1). In total, these three clones comprised 91%, 61%, 71% and 66%, of the respective collections from 1995 to 2003–2005. When data from 2003 to 2005 were combined, clone I was present in the largest proportion (36%) at the northern field sites, Elgin/Thurso (E/T). However, in other areas, clones C and J were more abundant than I. In Angus/Perthshire (A/P), J comprised 35%, and I and C, 17% and 14%, respectively. In Fife, clone C was predominant occurring at a frequency of 21%, whereas J was at 8% and I at 4% and, in Lothian/Borders (L/B), J was present in 45% of samples, with I making up 19% and C comprising 17%.
Figure 1. The clonal composition of field samples collected from four areas of Scotland from 1995 to 2005. The numbers of each clone is expressed as a percentage of the total number, which was calculated by dividing the number of individuals by the total number, n. Un, unclassified.
Download figure to PowerPoint
In 2003, clones D and E were collected, but only two aphids were identified as belonging to clone D (< 1%), whereas 16% of aphids comprised clone E (Fig. 1). Clones D and E were first identified in 2001 (Fenton et al., 2005) where they made up only a small proportion of the sample (7% and 4% respectively). Clone D was not collected again; however, clone E comprised 3% and 14% of the 2004 and 2005 samples, respectively.
Clone K was first collected in 2003 in Fife, and is unusual as it is a red–brown colour. When this clone was tested for insecticide resistance, it was found to be R3 and heterozygous for kdr. It was only collected once in 2003 and 2004 in A/P, but it was collected three times in 2005 at sites in E/T and A/P suggesting that, although rare, it is widespread in Scotland.
In 2003, 80 M. persicae were collected from fields close to Elgin (Fig. 1). Here, a new clone was found, designated as clone L, which made up 20% of the collection that year. In 2004 and 2005, it constituted 40% and 20% of the Elgin samples, respectively, making it the second most dominant clone in that area. Clone L was not collected in any of the other sample areas in any year, including Thurso, which is 120 km north of Elgin.
Clone F made up 5% of the collections in 1995 but it was not found again. DNA from clone F samples was tested for insecticide resistance status and the samples were all found to be R2/3 esterase and homozygous for kdr (kdr/kdr) (Foster et al., 2000; Anstead et al., 2004). Clone G, an insecticide sensitive clone, was found once in 1995 and once in 2003.
Three new MACE carrying clones were identified in 2003 and 2004 (H, M and N) and are the subject of a separate study (Kasprowicz et al., 2007). These supplemented the two MACE clones, A and B, identified by Fenton et al. (2005).
Distribution of clones on crop plants
Information gathered from unsprayed crops was used to examine the distribution of clones C, E, H, I, J and L on potato or brassica host plants (Fig. 4). Clones with numbers too small to be analysed individually (A, F, G and K) were combined into a single category. Chi-square analysis was used to compare the observed and expected distributions based on an assumption of equal proportions on both plant types: C, χ2 = 3.17, P > 0.05; E, χ2 = 10.35, P < 0.01; H, χ2 = 0.59, P > 0.20; I, χ2 = 0.30, P > 0.50; J, χ2 = 0.46, P > 0.20; L, χ2 = 1.85, P > 0.10; A + F + G + K, χ2 = 0.22, P > 0.50 (all with 1 d.f.). Therefore, there was no evidence that most clones were significantly associated with either host type. However, the increased proportion of clone E on brassica crops was highly significant, evidence that it demonstrated a preference for this particular host (Fig. 4).
Figure 4. The percentage distribution of abundant clones on potato and brassica crops at unsprayed sites. **P < 0.01.
Download figure to PowerPoint