Skip to main content
NCBI home page
Ecology and Evolution logoLink to Ecology and Evolution
. 2015 Sep 2;5(18):4098–4107. doi: 10.1002/ece3.1681

Queen execution increases relatedness among workers of the invasive Argentine ant, Linepithema humile

Maki N Inoue 1,✉, Fuminori Ito 2, Koichi Goka 3
PMCID: PMC4588641  PMID: 26445661

Abstract

Polygyny in social insects can greatly reduce within‐nest genetic relatedness. In polygynous ant species, potential rival queens in colonies with multiple queens are often executed by other queens, workers, or both. The Argentine ant, Linepithema humile, native to South America, forms a “supercolony” that is composed of a large number of nests and is considered to contribute to the ant's invasion success. Currently, four mutually antagonistic supercolonies are contiguously distributed within a small area of Japan. Here, we analyzed the genetic structure and relatedness within and among the four supercolonies using microsatellite markers to clarify how L. humile maintains its supercoloniality. The results of AMOVA and BASP, the FST values, and the existence of several private alleles indicated that the L. humile population in the Kobe area had a characteristic genetic structure. Within a given supercolony, there was significant genetic differentiation (FST) among workers collected in May and those collected in September. The significant deviation from Hardy–Weinberg equilibrium increased, and the relatedness among workers significantly increased from May to September in all supercolonies. This result suggested that the supercolonies replaced old queens with new ones during the reproductive season, thus supporting the plausibility of queen execution. From the perspective of kin selection, workers collectively eliminate queens, thereby increasing their own inclusive fitness. Restricted gene flow among supercolonies, together with mating with sib and queen execution, could help to maintain the unique social structure of L. humile, the distribution of which is expanding worldwide.

Keywords: Aggressiveness, biological invasion, Hymenoptera, microsatellite, social structure

Introduction

Eusocial Hymenoptera (ants, bees, and wasps) vary greatly in their social structure. The single queen structure (monogyny), which maximizes relatedness among offspring, is ancestral in the major lineages of eusocial insects (Boomsma 2007; Hughes et al. 2008), and the multiple queen structure (polygyny) has evolved secondarily. Workers in monogynous species help raise their full sibs and so gain inclusive fitness returns from their helping behavior. On the other hand, polygyny can greatly reduce within‐nest genetic relatedness, thus diluting the benefits of helping behavior. An increase in queen numbers is associated with a decrease in each queen's reproductive output (Sommer and Holldobler 1995). These queens compete for reproductive shares within a colony with limited resources. To increase their inclusive fitness, stingless bee workers of Melipona execute emerging virgin queens when the queen' numbers are very high (Jarau et al. 2009), being up to 16% (Wenseleers et al. 2004).

Among ant species, polygyny is a common social structure in which queens usually reduce their dispersal abilities and mate in, or close to, their natal nests (Bourke and Franks 1995). This is probably because mating among sibs increases relatedness between workers and the brood they rear within a colony (Crozier and Pamilo 1996). In polygynous ant species, potential rival queens in colonies newly established by multiple queens (i.e., colonies with pleometrosis) are often executed by other queens (Sommer and Holldobler 1995; Balas and Adams 1996a; Aron et al. 2009), the workers (Balas and Adams 1996b), or both (Helms et al. 2013). Queen execution could benefit not only queens themselves by eliminating rivals but also workers by killing extra queens and thus increasing exclusive fitness.

In places where it is introduced, the Argentine ant, Linepithema humile (Mayr), which is native to South America, forms a “supercolony” composed of a large number of nests and ranging over several thousand kilometers (Giraud et al. 2002). This social structure is considered to contribute to the ant's invasion success, because a supercolony, by reducing the costs associated with territoriality, may have high worker densities, thus promoting ecological dominance (Holway et al. 1998). On the other hand, a supercolony is also an evolutionary paradox for kin selection theory because supercolonies are genetically homogeneous entities, with the majority of workers having low relatedness to each other within colonies (Bourke and Franks 1995; Crozier and Pamilo 1996; Tsutsui and Case 2001; Helantera et al. 2009). Linepithema humile lacks nuptial flights (Passera and Keller 1990), and queens mate with related males within their natal nests (Markin 1970). At the beginning of the reproductive season, up to 90% of queens are usually killed by the workers (Keller et al. 1989; Reuter et al. 2001), and this queen execution may increase average worker relatedness among nestmates (Keller et al. 1989).

Linepithema humile was first reported in Japan in 1993 (Sugiyama 2000), and four mutually antagonistic supercolonies (Kobe A, Kobe B, Kobe C, and Japanese main), each of which is associated with one of four different mitochondrial haplotypes (Inoue et al. 2013), have been found in the city of Kobe, in western Japan (Murakami 2002; Okaue et al. 2007; Sunamura et al. 2009). Here, we used microsatellite markers to analyze genetic structure and relatedness within and among the four supercolonies to elucidate how L. humile maintains its supercoloniality.

Materials and Methods

Our studies were conducted in 2009 in the four L. humile populations established in the Kobe area: the KA (Kobe A) and KB (Kobe B) supercolonies at Port Island, and the KC (Kobe C) and JM (Japanese main) supercolonies at the Maya Wharf (Fig. 1). We collected 11–20 workers from each of 19 nests in May and September 2009 for genetic analysis. Sexual production of L. humile generally occurs in the spring and early summer (M. N. Inoue et al. unpubl. data): May is just before reproduction begins and September is when the first workers are produced by the new queens. We also collected workers from 17 nests (excluding KA4 and KB5) to perform aggression tests between workers from different nests. The distance between the nests ranged from 30 to 380 m (mean ± SE, 177.50 ± 69.6 m) at the Maya Wharf and 40–1270 m (506.00 ± 29.5 m) at the Port Island site.

Figure 1.

Figure 1

Open in a new tab

Map of the study populations of Linepithema humile in the Kobe area. Population codes: KA, Kobe A; KB, Kobe B; KC, Kobe C; and JM, Japanese main.

Aggression tests

Intercolony aggression tests between workers were performed in May and September to confirm that workers sampled from each nest belonged to the expected supercolony. We randomly selected one worker from one of the nests from each supercolony and placed it in a plastic dish (diameter: 4 cm). One worker from a nest from the adjacent supercolony was then added and the two were observed for 5 min. To quantify the workers' behavior, we scored each contact using a 0–4 scale (modified from the work of Suarez et al. 1999; see also Inoue et al. 2013), as follows: 0 = ignoring (physical contact between individuals in which neither ant showed any interest in the other ant); 1 = avoidance (one or both ants retreated after contact) or antennation (repeated tapping of the antennae somewhere on the other ant); 2 = dorsal flexion (abdomen pointed toward the other ant in a threatening posture); 3 = aggression (biting the other ant); and 4 = fighting (prolonged aggression, resulting in severe injury or death). Six tests were conducted for each pair of nests using different individuals each time.

DNA extraction and PCR amplification

We assessed the genotypes of the ants in each nest using microsatellite loci. DNA was extracted from individual workers according to the method described by Goka et al. (2000). After the application of 60 μL of lysis buffer (1 mg/mK proteinase K, 0.01 mol/L NaCl, 0.1 mol/L EDTA, 0.01 mol/L Tris–HCl [pH 8.0], 0.5% Nonidet P‐40), each worker was homogenized at 50°C for 60 min and then at 94°C for 10 min. We then diluted 30 μL of the homogenate with 270 μL of TE buffer (0.001 mol/L EDTA, 0.001 mol/L Tris–HCl [pH 8.0]) and stored it at −20°C. Amplification reactions were performed in 25 μL volumes consisting of 0.5 μL of template DNA, 0.2 mmol/L dNTP, 4 mmol/L MgCl2, 0.5 units of Taq DNA polymerase (Amplitaq Gold; Applied Biosystems, Foster City, CA), and 0.4 μmol/L of each primer. Samples were amplified using PCR under the following conditions: initial denaturing at 95°C for 10 min followed by 30 cycles of denaturing at 94°C for 10 sec, annealing at 55°C for 20 sec, and extension at 72°C at 20 sec, with a final extension step at 72°C for 7 min. Seven microsatellite loci were analyzed: Lhum‐3, Lhum‐18, Lhum‐19, Lhum‐28, and Lhum‐52 (Krieger and Keller 1999); and Lihu‐M1 and Lihu‐S3 (Tsutsui et al. 2000). Primers were labeled with fluorescent dyes, and PCR products were visualized with an ABI 3770 sequencer (Applied Biosystems).

Statistical analysis

Data are expressed as means ± SD. To compare the levels of aggression between workers at each site, we employed a generalized linear mixed‐effect model (GLMM; Venables and Ripley 2002 ) with the assumption of a normal distribution of errors with “pair” (i.e., workers of KA and KB and workers of KC and JM) as the fixed effect and “nest” as the random effect, using R version 3.0.1 (R Development Core Team 2013) and the nlme libraries (Pinheiro and Bates 2000).

Random mating (i.e., agreement with the conditions for Hardy–Weinberg equilibrium, HWE) within each supercolony was tested using Arlequin version 3.5.1.3 (Excoffier and Lischer 2010). An unbiased estimate of the exact P value for each locus was computed using the MCMC (Markov chain Monte Carlo) method with a forecast chain length of 100,000 steps and 10,000 dememorization steps. Linkage disequilibrium between loci in each population was conducted with MCMC with 10,000 dememorization steps, 1000 batches, and 500 iterations per batch using Genpop on the Web (http://genepop.curtin.edu.au/, accessed on 25 December 2013). Null allele frequencies were calculated for each locus in accordance with the method of Brookfield (1996) using Micro‐Checker version 2.2.3 (van Oosterhout et al. 2004). We also performed analysis of molecular variance (AMOVA) and Mantel tests and calculated pairwise F ST (genetic differentiation) values and three other genetic diversity measures – the number of observed alleles per locus (A O), observed heterozygosity (H O), and expected heterozygosity (H E) –using Arlequin, and allelic richness (A R) using FSTAT version 2.9.3.2 (Goudet 2002). Private alleles were identified by pairwise comparisons among supercolonies. The significance of the AMOVA output was tested using 1000 permutations. All pairwise F ST values were calculated at two levels (nest and supercolony). Pairwise F ST values between nests were plotted against the geographic distance between the nests to test for the possibility of genetic isolation by distance. The significance of the regression coefficients was tested using the Mantel test with 10,000 permutations. Pairwise relatedness indices (r xy; Queller and Goodnight 1989) between individuals within each supercolony in each season (May and September) were calculated using KINGROUP (Konovalov et al. 2004). The statistical differences between the r xy values in May and September were determined using a two‐tailed Student's t‐test in R.

To confirm the assignment of individuals to each of the four supercolonies, we used the Bayesian clustering method implemented in STRUCTURE 2.1 (Pritchard et al. 2000). In this analysis, we assumed that all individuals belonged to a hypothetical single population. Having sampled a total of 38 nests from the four supercolonies (all 19 nests in May and again in September), we tested K values (the number of distinct groups) from 1 to 20 to identify the best fit. For each K, we performed 20 runs with 100,000 replications of the MCMC procedure following a burn‐in period of 100,000 replications. We then calculated the average of the 10 highest likelihood values [lnP(D)] for each K, and we plotted the lnP(D) values against the K value to identify the best K. The K value was also calculated using the ΔK statistic, which is based on the rate of change of the logarithm of probability between successive analyses of K (Evanno et al. 2005). The group‐level Bayesian analysis in BSPS 6.0 (Bayesian Analysis of genetic Population Structure [BAPS], Corander et al. 2008) was also conducted with the maximum number of clusters set to between 2 and 10 according to the results obtained from analysis of the lnP(D) values analyzed by STRUCTURE. The results of STRUCTURE tend to be conservative in terms of the number of clusters detected as ancestral information related to the introduction history of a species; BAPS performs better in clustering together populations with recent gene flow (Díez‐del‐Molino et al. 2013).

Results

Aggression

Aggressive behavior was observed between workers from adjacent supercolony pairs. The average levels of aggression between the KA and KB supercolonies (3.65 ± 0.33 in both May and September) were significantly larger than those of the KC and JM supercolonies (GLMM, estimate = −0.67, SE = 0.16, t = −4.19, P < 0.0001 in May; GLMM, estimate = −0.50, SE = 0.17, t = −2.81, P = 0.0081 in September).

Genetic diversity and structure

At seven microsatellite loci, we assessed the genotypes of 717 individuals collected from the four supercolonies in May and September (for a total of eight population–month combinations) (Table 1). We detected 59 alleles in total. Of the 56 tests, 14 (25%) showed significant deviation from HWE (P < 0.05) in L. humile: The number of tests that deviated significantly from the expectation increased in the supercolonies was greater in September (nine) than in May (five). Significant linkage disequilibrium was found between Lhum18 and LihM1 (P < 0.05) after Bonferroni correction. The presence of null alleles was suggested at five loci: Lhum3, Lhum18, Lhum19, Lhum28, and LihM1. The mean estimated null allele frequency ranged from 0.004 (Lhum‐52) to 0.180 (LihuM1); across all seven loci, it was 0.079.

Table 1.

The expected heterozygosity (H E), observed number of alleles per locus (A O), allelic richness (A R), and P value for Hardy–Weinberg equilibrium (P HW) for the four supercolonies in the Kobe area, as determined from individuals collected in May and September 2009. See Fig. 1 for colony abbreviations

Month Supercolony n Lhum‐3 Lhum‐18 Lhum‐19 Lhum‐28 Lhum‐52 Lihu‐M1 Lihu‐S3 Mean SD
May KA 120 H E 0.76 0.70 0.66 0.61 0.45 0.16 0.51 0.55 0.20
H O 0.46 0.62 0.19 0.49 0.48 0.07 0.50 0.40 0.19
A O 6 4 6 5 4 8 4 5.29 1.50
A R 6.00 4.00 5.48 4.49 3.22 6.16 3.45 4.68 1.21
P HW 0.000 0.000 0.000 0.009 0.008 0.000 0.008
KB 117 H E 0.83 0.63 0.72 0.14 0.00 0.70 0.04 0.44 0.36
H O 0.74 0.52 0.66 0.14 0.00 0.44 0.03 0.36 0.30
A O 9 4 4 2 1 5 2 3.86 2.67
A R 8.43 4.00 4.00 2.00 1.00 5.00 1.98 3.77 2.50
P HW 0.000 0.000 0.049 0.508 0.000 0.000 0.045
KC 80 H E 0.69 0.74 0.68 0.67 0.77 0.66 0.65 0.69 0.05
H O 0.59 0.61 0.43 0.58 0.73 0.43 0.68 0.58 0.11
A O 5 7 7 5 6 8 4 6.00 1.41
A R 4.73 6.93 6.85 4.98 5.73 7.72 5.94 6.12 1.09
P HW 0.000 0.000 0.000 0.118 0.056 0.000 0.023
JM 60 H E 0.67 0.53 0.69 0.07 0.50 0.13 0.44 0.43 0.25
H O 0.32 0.38 0.60 0.03 0.53 0.00 0.50 0.34 0.24
A O 4 4 5 3 2 2 3 3.29 1.11
A R 4.00 4.00 5.00 3.00 2.00 2.00 2.97 3.28 1.11
P HW 0.000 0.000 0.003 0.000 0.612 0.000 0.572
September KA 120 H E 0.54 0.67 0.66 0.61 0.44 0.17 0.50 0.51 0.17
H O 0.55 0.82 0.31 0.46 0.44 0.04 0.48 0.44 0.23
A O 6 4 6 4 4 4 2 4.29 1.38
A R 5.21 4.00 5.51 4.00 2.97 3.93 2.00 3.94 1.21
P HW 0.000 0.000 0.000 0.001 0.010 0.000 0.717
KB 89 H E 0.70 0.72 0.72 0.16 0.00 0.52 0.05 0.41 0.33
H O 0.69 0.63 0.65 0.17 0.00 0.36 0.05 0.36 0.30
A O 7 5 5 2 1 4 2 3.71 2.14
A R 6.61 4.88 4.65 2.00 1.00 3.76 1.99 3.56 1.99
P HW 0.000 0.000 0.053 1.000 0.000 0.000 1.000
KC 71 H E 0.62 0.62 0.69 0.70 0.72 0.72 0.61 0.67 0.05
H O 0.62 0.61 0.29 0.54 0.80 0.55 0.68 0.58 0.16
A O 4 7 7 4 5 9 4 5.71 1.98
A R 4.00 6.63 6.83 4.00 4.97 8.78 6.60 5.97 1.74
P HW 0.000 0.178 0.000 0.028 0.016 0.000 0.072
JM 60 H E 0.39 0.61 0.66 0.00 0.52 0.03 0.35 0.37 0.26
H O 0.42 0.70 0.67 0.00 0.50 0.03 0.35 0.38 0.28
A O 4 5 4 1 3 2 2 3.00 1.41
A R 3.93 5.00 3.97 1.00 3.00 2.00 2.00 2.99 1.40
P HW 0.011 0.000 0.826 0.000 0.014 1.000 1.000 　 　
Open in a new tab

Across loci, the expected heterozygosity (H E) ranged from 0.00 to 0.83 (mean = 0.51 ± 0.25), and the observed heterozygosity (H O) ranged from 0.00 to 0.82 (mean = 0.42 ± 0.24). The number of alleles per locus (A O) for all samples ranged from 1 to 9, and the mean allelic richness (A R) ranged from 2.99 to 6.12. Mean A R was highest in the KC supercolony and lowest in the JM supercolony in both months and decreased in all supercolonies between May and September. In all, 24 private alleles were detected across the seven loci, with at least one private allele per locus: Lhum‐3 produced three; Lhum‐18, seven; Lhum‐19, four; Lhum‐28, three; Lhum‐52, one; Lihu‐M1, four; and Lihu‐S3, two. All supercolonies had two or more private alleles: Five were detected in KA; four in KB; 13 in KC; and two in JM.

The mean relatedness values (r xy) of worker nestmates within each supercolony ranged from 0.208 to 0.407 in May and from 0.307 to 0.584 in September (Table 2); the r xy values within each supercolony were significantly lower in May than in September (t‐test, P < 0.0001).

Table 2.

Mean and standard deviation (SD) of relatedness (r xy) values for nestmate workers collected from the four supercolonies in May and September 2009. See Fig. 1 for colony abbreviations

Month Supercolony Mean r xy SD
May KA 0.208 0.260
KB 0.395 0.208
KC 0.290 0.202
JM 0.407 0.253
September KA 0.318 0.261
KB 0.449 0.209
KC 0.307 0.190
JM 0.584 0.200
Open in a new tab

The F ST values ranged from 0.172 (KB vs. JM) to 0.370 (KC vs. JM) in May and from 0.191 (KA vs. KB) to 0.416 (KC vs. JM) in September among the four supercolonies; all values were significant (P < 0.0001, Table 3). The F ST values also differed significantly among workers collected in May and those collected in September within each given supercolony, ranging from 0.027 (KB vs. KB) to 0.068 (JM vs. JM) (P < 0.0001, Table 3). AMOVA detected a significant population structure for the microsatellite data, with small but significant components of variance within (F SC) and between (F CT) supercolonies (F SC = 1.78, %V = 73.80, P < 0.0001; F CT = 0.63, %V = 26.20, P < 0.0001). The relationship between the pairwise F ST and the geographic distance between nests was not significant within each supercolony: KA, r 2 = 0.057, P = 0.316 in May and r 2 = 0.093, P = 0.142 in September (Fig. 2A and B); KB, r 2 = 0.060, P = 0.193 in May and r 2 = −0.001, P = 0.339 in September (Fig. 2A and B); KC, r 2 = −0.245, P = 0.912 in May and r 2 = −0.569, P = 0.051 in September (Fig. 2C and D); and JM, r 2 = 0.949, P = 0.900 in May and r 2 = −0.786, P = 0.788 in September (Fig. 2C and D).

Table 3.

Pairwise F ST comparisons between samples from the four supercolonies in the Kobe area, collected in May and September 2009. All values were significant (P < 0.0001). Values for the comparison of a supercolony in May with the same supercolony in September are shown in the bottom half of the table. See Fig. 1 for colony abbreviations

Month Supercolony May September
KA KB KC JM KA KB KC JM
May KA
KB 0.188
KC 0.293 0.393
JM 0.203 0.172 0.370
September KA 0.040 0.224 0.329 0.234
KB 0.191 0.027 0.403 0.191 0.215
KC 0.292 0.388 0.051 0.382 0.306 0.393
JM 0.241 0.222 0.414 0.068 0.244 0.195 0.416
Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Pairwise genetic distance FST values between workers from different nests plotted against the geographic distance between nests: The KA and KB supercolonies were collected in (A) May and (B) September 2009 at Port Island, and the KC and JM supercolonies were collected in (C) May and (D) September 2009 at Maya Wharf. JM, Japanese main; KA, Kobe A; KB, Kobe B; KC, Kobe C.

Figure 3 shows the relationship between lnP(D) and K. The lnP(D) values increased with increasing K until they reached a plateau at about K = 9. If the supercolonies had formed a significant genetic barrier to the flow of nuclear genes, we would have expected four population clusters to be revealed in this analysis; therefore, the lnP(D) values would have approached a maximum at K = 4. However, the curve failed to produce a clear peak at any value of K and the STRUCTURE analysis did not detect any signs of population structure in L. humile. The characteristics of the curve were not altered by extending the burn‐in period or by increasing the number of MCMC iterations. The log‐likelihood curve was further tested using the ΔK statistic, but this statistic also did not produce a distinct peak that would have been correlated with the most likely value of K. All individuals were assigned to one of the respective clusters with a probability of >82% at K = 4 according to the number of supercolonies using STRUCTURE (Fig. 4A). In contrast, BAPS identified eight homogenous units within the study area, indicating that each supercolony represented genetic differentiation between the seasons (Fig. 4B).

Figure 3.

Figure 3

Open in a new tab

Plot of log‐likelihood values [lnP(D)] as a function of K indicating the number of distinct groups for Linepithema humile sampled from 19 nests in the Kobe area in both May and September 2009, for a total of 38 nest–month combinations. All data were obtained from analysis of the microsatellite DNA loci using 20 runs with STRUCTURE. K values tested ranged from 1 to 20, with a burn‐in of 100,000 and with 100,000 Markov chain Monte Carlo iterations per value of K.

Figure 4.

Figure 4

Open in a new tab

Bayesian analyses of population structure. Analyses were performed with (A) STRUCTURE and (B) BAPS among the Linepithema humile supercolonies: (A) Each individual is represented as a vertical bar partitioned into segments of different color according to the proportion of the genome belonging to each of the four identified clusters (K = 4); each sampling is a different color according to the cluster to which it belongs.

Discussion

In our analysis using microsatellite genotypes, the results of AMOVA and BAPS, the F ST values, and the existence of several private alleles indicated that the L. humile population in the Kobe area was characterized by a pronounced genetic structure. As in previous studies conducted in Europe (Giraud et al. 2002; Jaquiery et al. 2005), North America (Tsutsui and Case 2001; Thomas et al. 2006), and Japan (Sunamura et al. 2009), our findings support the notion that gene flow among supercolonies is largely restricted and that mutually antagonistic supercolonies remain genetically differentiated. On the other hand, our STRUCTURE results did not detect clear genetic units; instead, they demonstrated that some workers of the KA supercolony were assigned to the adjacent KB supercolony with low probability (Fig. 4A). This suggested that gene flow occurred from the KB supercolony to the KA supercolony. In L. humile, only the males disperse, whereas the queens mate and stay within or close to their natal nests (Passera and Keller 1990). During the early reproductive period, under laboratory conditions, foreign males are not attacked by workers (M.N. Inoue, pers. obs.); instead, they are adopted by, and successfully mate with, resident queens (Passera and Keller 1994). However, at the end of reproduction, males are aggressively attacked by the workers (Sunamura et al. 2011). Therefore, male‐mediated gene flow might occur between the supercolonies, but only during the early reproductive period.

Within each supercolony, the genetic differentiation (F ST) between nests was low; this is also consistent with previous studies both in native populations in Argentina (Pedersen et al. 2006) and in areas where the ant has been introduced (Giraud et al. 2002). In addition, there was no evidence for a correlation between genetic differentiation and the geographic distance among the nests of the four supercolonies, suggesting that isolation by distance does not appear to be occurring at a small spatial scale; this supports the findings of Tsutsui and Case (2001). On the other hand, we demonstrated that significant deviation from HWE was grater in September than in May in all supercolonies. This result suggested the occurrence of nonrandom mating, probably because queens mate with their sibs within a nest, despite the above‐mentioned possibility of gene flow among supercolonies. The significant genetic differentiation (F ST) between workers collected in May and those collected in September within a given supercolony suggested that the supercolonies replaced old queens with new ones after the reproductive season, supporting the plausibility of queen execution. Queen execution has been reported in two introduced populations – in the United States (Markin 1970) and in France (Keller et al. 1989) – but not in native populations. Most executed queens were <1 year old, suggesting that queen age, like weight and fecundity (Keller et al. 1989), does not play an important role in the choice of queen for execution. From the perspective of kin selection, one hypothesis proposed by Keller et al. (1989) is that workers collectively eliminate queens, to which they are less related, thereby increasing their own inclusive fitness. Reuter et al. (2001), however, reported that the workers in a nest were, on average, not significantly less related to executed queens than to surviving ones. We revealed here that relatedness among workers increased significantly from May to September in all supercolonies, probably as a result of queen execution, supporting the hypothesis of Keller et al. (1989). Our results also showed that genetic diversity (allelic richness) decreased significantly from May to September, indicating that the occurrence of genetic drift due to queen execution may contribute to genetic differentiation by changing the gene frequencies within a supercolony.

The genetic differentiation among supercolonies and the genetic homogeneity within each supercolony in this recently established Japanese population suggest that each introduced supercolony reflects the genetic constitution of the founder colony rather than the mixing of several mutually antagonistic supercolonies. Thus, the colony structure does not differ between its area of origin and area of introduction. Among the four supercolonies, the JM and KB supercolonies were characterized by lower genetic diversity and higher relatedness among workers than were the KA and KC supercolonies (see Tables 2, 3). First, these differences among supercolonies may arise from differences in invasion history. The JM supercolony represents the dominates across Europe, North America, Australasia, and Japan on each continent or island (Tsutsui et al. 2000; Corin et al. 2007; Giraud et al. 2002; Inoue et al. 2013; Suhr et al. 2009). The secondary large KB supercolony has also been found in some populations in Japan and the United States, whereas the minor KA and KC supercolonies are locally distributed within their introduced ranges (Inoue et al. 2013). The two dominant supercolonies are likely to have been transferred stepwise from their native range to a new region and then on to other new regions and to have originated from established populations. In the areas to which they are introduced, they may have little opportunity for gene flow with other supercolonies because of their high levels of dominance. Therefore, low genetic diversity as well as high relatedness due to mating with sibs and queen execution can be found in these two dominant supercolonies. On the other hand, the two minor supercolonies may not have had repeated invasion histories and thus maintain relatively high genetic diversity but low relatedness despite queen execution. Second, the differences in genetic structure among the supercolonies may reflect those of the source populations in the native ranges, although variations in relatedness among nestmates have not been found in the native supercolonies (Vogel et al. 2009).

In conclusion, restricted gene flow among supercolonies, mating with sibs, and queen execution could help to maintain the unique social structure of L. humile, the distribution of which is expanding worldwide.

Conflict of Interest

None declared.

Acknowledgments

We thank A. Tominaga, T. Okamoto, and S. Tokoro for their help in the field and their valuable advice, and K. Suzuki and A. Miyamoto for their help in the laboratory. We also thank T. Kishimoto, M. Kobayashi, K. Nishina, and S. Moriguchi for their helpful comments. This study was supported by the Global Environment Research Fund (F‐081, Leader: K. Goka) of the Ministry of the Environment and by a Grant‐in‐Aid for Scientific Research (C) from KAKENHI (22570031) to M. Inoue.

Ecology and Evolution 2015; 5(18): 4098–4107

References

  1. Aron, S. , Steinhauer N., and Fournier D.. 2009. Influence of queen phenotype, investment and maternity apportionment on the outcome of fights in cooperative foundations of the ant Lasius niger . Anim. Behav. 77:1067–1074. [Google Scholar]
  2. Balas, M. T. , and Adams E. S.. 1996a. The dissolution of cooperative groups: mechanisms of queen mortality in incipient fire ant colonies. Behav. Ecol. Sociobiol. 38:391–399. [Google Scholar]
  3. Balas, M. T. , and Adams E. S.. 1996b. Nestmate discrimination and competition in incipient colonies of fire ants. Anim. Behav. 51:49–59. [Google Scholar]
  4. Boomsma, J. J. 2007. Kin selection versus sexual selection: why the ends do not meet. Curr. Biol. 17:R673–R683. [DOI] [PubMed] [Google Scholar]
  5. Bourke, A. F. G. , and Franks N. R.. 1995. Social evolution in ants. Princeton Univ. Press, Princeton, NJ. [Google Scholar]
  6. Brookfield, J. F. Y. 1996. A simple new method for estimating null allele frequency from heterozygote deficiency. Mol. Ecol. 5:453–455. [DOI] [PubMed] [Google Scholar]
  7. Corander, J. , Marttinen P., Sirén J., and Tang J.. 2008. Enhanced Bayesian modelling in BAPS software for learning genetic structures of populations. BMC Bioinformatics 9:539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Corin, S. E. , Abbott K. L., Ritchie P. A., and Lester P. J.. 2007. Large scale unicoloniality: the population and colony structure of the invasive Argentine ant (Linepithema humile) in New Zealand. Insectes Soc. 54:275–282. [Google Scholar]
  9. Crozier, R. H. , and Pamilo P.. 1996. Evolution of social insect colonies: sex allocation and kin selection. Oxford Univ. Press, Oxford. [Google Scholar]
  10. Díez‐del‐Molino, D. , Carmona‐Catot G., Araguas R., Vidal O., Sanz N., García‐Berthou E., et al. 2013. Gene flow and maintenance of genetic diversity in invasive mosquitofish (Gambusia holbrooki). PLoS ONE 8:e82501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Evanno, G. , Regnaut S., and Goudet J.. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol. Ecol. 14:2611–2620. [DOI] [PubMed] [Google Scholar]
  12. Excoffier, L. , and Lischer H. E.. 2010. Arlequin Suite Ver. 3.5: a new series of programs to perform population genetic analyses under Linux and Windows. Mol. Ecol. Resour. 10:564–567. [DOI] [PubMed] [Google Scholar]
  13. Giraud, T. , Pedersen J. S., and Keller L.. 2002. Evolution of supercolonies: the Argentine ants of southern Europe. Proc. Natl Acad. Sci. USA 99:6075–6079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Goka, K. , Okabe K., Niwa S., and Yoneda M.. 2000. Parasitic mite infestation in introduced colonies of European bumblebees, Bombus terrestris . Jpn. J. Appl. Entomol. Zool. 44:47–50. [Google Scholar]
  15. Goudet, J. 2002. Fstat 2.9.3.2. URL: http://www2.unil.ch/popgen/softwares/fstat.htm.
  16. Helantera, H. , Strassmann J. E., Carrillo J., and Queller D. C.. 2009. Unicolonial ants: where do they come from, what are they and where are they going? Trends Ecol. Evol. 24:341–349. [DOI] [PubMed] [Google Scholar]
  17. Helms, K. R. , Newman N. J., and Cahan S. H.. 2013. Regional variation in queen and worker aggression in incipient colonies of the desert ant Messor pergandei . Behav. Ecol. Sociobiol. 67:1563–1573. [Google Scholar]
  18. Holway, D. A. , Suarez A. V., and Case T. J.. 1998. Loss of intraspecific aggression in the success of a widespread invasive social insect. Science 282:949–952. [DOI] [PubMed] [Google Scholar]
  19. Hughes, W. O. H. , Oldroyd B. P., Beekman M., and Ratnieks F. L. W.. 2008. Ancestral monogamy shows kin selection is key to the evolution of eusociality. Science 320:1213–1216. [DOI] [PubMed] [Google Scholar]
  20. Inoue, M. N. , Sunamura E., Suhr E. L., Ito F., Tatsuki F., and Goka K.. 2013. Recent range expansion of the Argentine ant in Japan. Divers. Distrib. 19:29–37. [Google Scholar]
  21. Jaquiery, J. , Vogel V., and Keller L.. 2005. Multilevel genetic analyses of two European supercolonies of the Argentine ant, Linepithema humile . Mol. Ecol. 14:589–598. [DOI] [PubMed] [Google Scholar]
  22. Jarau, S. , Van Veen J. W., Aguilar I., and Ayasse M.. 2009. Virgin queen execution in the stingless bee Melipona beecheii: the sign stimulus for worker attacks. Apidologie 40:496–507. [Google Scholar]
  23. Keller, L. , Passera L., and Suzzoni J. P.. 1989. Queen execution in the Argentine ant, Iridomyrmex humilis . Physiol. Entomol. 14:157–163. [Google Scholar]
  24. Konovalov, D. A. , Manning C., and Henshaw M. T.. 2004. KINGROUP: a program for pedigree relationship reconstruction and kin group assignments using genetic markers. Mol. Ecol. Notes 4:779–782. [Google Scholar]
  25. Krieger, M. J. B. , and Keller L.. 1999. Low polymorphism at 19 microsatellite loci in a French population of Argentine ants (Linepithema humile). Mol. Ecol. 8:1078–1080. [Google Scholar]
  26. Markin, G. 1970. The seasonal life cycle of the Argentine ant, Iridomyrmex humilis (Hymenoptera, Formicidae), in southern California. Ann. Entomol. Soc. Am. 63:1238–1242. [Google Scholar]
  27. Murakami, K. 2002. Exotic ants in portIsland, Kobe City. ARI 26:45–46 (in Japanese). [Google Scholar]
  28. Okaue, M. , Yamamoto K., Touyama Y., et al. 2007. Distribution of the Argentine ant, Linepithema humile, along the Seto Inland Sea, western Japan: result of surveys in 2003–2005. Entomol. Sci. 10:337–342. [Google Scholar]
  29. van Oosterhout, C. , Hutchinson W. F., Wills D. P. M., and Shipley P.. 2004. MICRO‐CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4:535–538. [Google Scholar]
  30. Passera, L. , and Keller L.. 1990. Loss of mating flight and shift in the pattern of carbohydrate storage in sexuals of ants (Hymenoptera; Formicidae). J. Comp. Physiol. B Biochem. Syst. Environ. Physiol. 160:207–211. [Google Scholar]
  31. Passera, L. , and Keller L.. 1994. Mate availability and male dispersal in the Argentine ant Linepithema humile (Mayr) (=Iridomyrmex humilis). Anim. Behav. 48:361–369. [Google Scholar]
  32. Pedersen, J. S. , Krieger M. J. B., Vogel V., Giraud T., and Keller L.. 2006. Native supercolonies of unrelated individuals in the invasive Argentine ant. Evolution 60:782–791. [PubMed] [Google Scholar]
  33. Pinheiro, J. C. , and Bates D. M.. 2000. Mixed‐effects models in S and S‐plus. Springer Verlag, New York, NY. [Google Scholar]
  34. Pritchard, J. K. , Stephens M., and Donnelly P.. 2000. Inference of population structure using multilocus genotype data. Genetics 155:945–959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Queller, D. C. , and Goodnight K. F.. 1989. Estimating relatedness using genetic markers. Evolution 43:258–275. [DOI] [PubMed] [Google Scholar]
  36. R Development Core Team . 2013. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Google Scholar]
  37. Reuter, M. , Balloux F., Lehmann L., and Keller L.. 2001. Kin structure and queen execution in the Argentine ant Linepithema humile . J. Evol. Biol. 14:954–958. [Google Scholar]
  38. Sommer, K. , and Holldobler B.. 1995. Colony founding by queen association and determinants of reduction in queen number in the ant Lasius niger . Anim. Behav. 50:287–294. [Google Scholar]
  39. Suarez, A. V. , Tsutsui N. D., Holway D. A., and Case T. J.. 1999. Behavioral and genetic differentiation between native and introduced populations of the argentine ant. Biological Invasions 1:43–53. [Google Scholar]
  40. Sugiyama, T. 2000. Invasion of Argentine ant, Linepithema humile, into Hiroshima Prefecture, Japan. Jpn. J. Appl. Entomol. Zool. 44:127–129. [Google Scholar]
  41. Suhr, E. L. , McKechnie S. W., and O'Dowd D. J.. 2009. Genetic and behavioural evidence for a city‐wide supercolony of the invasive Argentine ant Linepithema humile (Mayr) (Hymenoptera: Formicidae) in southeastern Australia. Aust. J. Entomol. 48:79–83. [Google Scholar]
  42. Sunamura, E. , Hatsumi S., Karino S., et al. 2009. Four mutually incompatible Argentine ant supercolonies in Japan: inferring invasion history of introduced Argentine ants from their social structure. Biol. Invasions 11:2329–2339. [Google Scholar]
  43. Sunamura, E. , Hoshizaki S., Sakamoto H., et al. 2011. Workers select mates for queens: a possible mechanism of gene flow restriction between supercolonies of the invasive Argentine ant. Naturwissenschaften 98:361–368. [DOI] [PubMed] [Google Scholar]
  44. Thomas, M. L. , Payne‐Makrisa C. M., Suarez A. V., Tsutsui N. D., and Holway D. A.. 2006. When supercolonies collide: territorial aggression in an invasive and unicolonial social insect. Mol. Ecol. 15:4303–4315. [DOI] [PubMed] [Google Scholar]
  45. Tsutsui, N. D. , and Case T. J.. 2001. Population genetics and colony structure of the argentine ant (Linepithema humile) in its native and introduced ranges. Evolution 55:976–985. [DOI] [PubMed] [Google Scholar]
  46. Tsutsui, N. D. , Suarez A. V., Holway D. A., and Case T. J.. 2000. Reduced genetic variation and the success of an invasive species. Proc. Natl Acad. Sci. USA 97:5948–5953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Venables, W. N. , and Ripley B. D.. 2002. Modern applied statistics with S. Springer, New York, NY. [Google Scholar]
  48. Vogel, V. , Pedersen J. S., d'Ettorre P., Lehmann L., and Keller L.. 2009. Dynamics and genetic structure of Argentine ant supercolonies in their native range. Evolution 63:1377–1671. [DOI] [PubMed] [Google Scholar]
  49. Wenseleers, T. , Hart A. G., Ratnieks F. L. W., and Quezada‐Euan J. J. G.. 2004. Queen execution and caste conflict in the stingless bee Melipona beecheii . Ethology 110:725–736. [Google Scholar]

Articles from Ecology and Evolution are provided here courtesy of Wiley

RESOURCES