SEARCH

SEARCH BY CITATION

Keywords:

  • arrhenotoky;
  • Neochrysocharis formosa;
  • Rickettsia;
  • sympatric speciation;
  • thelytoky;
  • wasp

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Sympatric speciation is strictly defined as the emergence of two species from a population in which mating has been random with respect to the place of birth of the mating partners. Mathematical models have shown that sympatric speciation is possible, but very few examples have been documented in nature. In this article, we demonstrate that arrhenotokous and thelytokous strains of a parasitic wasp, Neochrysocharis formosa, speciated sympatrically through infection by a symbiotic bacterium Rickettsia for the following reasons: First, Rickettsia infection was detected in all of the thelytokous strains collected throughout Japan. Second, the arrhenotokous and thelytokous strains have been collected sympatrically. Third, crossing experiments between the two strains did not result in fertilized offspring. In addition, the two strains were genetically isolated at the nuclear and mitochondrial genes. Fourth, the two strains showed a sister relationship in nuclear 28S rRNA gene. Finally, thelytokous females treated with antibiotics produced Rickettsia-free male offspring of the same reproductive form as arrhenotokous females indicating that the thelytokous strain could have speciated sympatrically from an individual of the arrhenotokous strain.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Speciation, the division of populations into evolutionarily independent units, involves genetic separation and phenotypic differentiation. Genetic divergence following geographic isolation gives rise to allopatric speciation: “the conceptual rationale is simply that, given enough time, speciation is an inevitable consequence of populations evolving in allopatry” (Turelli et al., 2001). Numerous empirical examples support this uncontroversial scenario (Coyne & Orr, 2004). In theory, however, populations can become genetically isolated without geographic separation, resulting in sympatric speciation, a much more contentious model. Sympatric speciation is more strictly defined as the emergence of two species from a population in which mating has been random with respect to the place of birth of the mating partners (Gavrilets, 2003).

Mathematical models have shown that sympatric speciation is possible (Dieckmann & Doebeli, 1999; Higashi et al., 1999; Kondrashov & Kondrashov, 1999; Tregenza & Butlin, 1999; Gavrilets, 2003), but very few examples have been documented in nature (Schliewen et al., 1994; Filchak et al., 2000). Cichlid fish seem to have radiated sympatrically in African crater lakes. Molecular phylogenetic analyses show that the fish species in each lake share a common ancestor, with sexual selection and ecology possibly driving speciation (Meyer et al., 1990; Barluenga et al., 2006). In host race examples, races of apple and hawthorn maggot have shifted to different hosts in sympatry and differ in reproductive behaviour and breeding time (Bush, 1994), and a genetic study of African indigobirds, which are host-specific brood parasites, showed that they might have recently speciated sympatrically after new hosts were colonized (Sorenson et al., 2003). In another example, two palm species found on an isolated Australian island segregate according to the acidity of the soil on which they are found (Savolainen et al., 2006). In an example termed as allochronic speciation, it was shown that small genetic differences between two lacewing species of Chrysopa caused temporal asynchrony in their life cycles that prevented reproduction and resulted in sympatric speciation (Tauber & Tauber, 1977; Tauber et al., 1977). Claims of sympatric speciation must demonstrate species sympatry, sister relationships, reproductive isolation and that an earlier allopatric phase is highly unlikely (Coyne & Orr, 2004).

Intracellular bacteria Wolbachia are known to manipulate the reproductive system of their hosts to their own benefit (see Werren et al., 2008 for reviews). Most commonly, Wolbachia cause sperm–egg incompatibilities, known as cytoplasmic incompatibility (Telschow et al., 2007). In concert with geographic or genetic barriers to gene flow, such cytoplasmic incompatibility (CI) could promote the evolution of reproductive isolation, a crucial component in speciation. In parasitic wasps of the genus Nasonia, for example, it was shown that, under laboratory conditions, both strains (Perrot-Minnot et al., 1996) and species (Bordenstein et al., 2001) can become fully reproductively isolated when they are infected with different Wolbachia strains. Parthenogenesis-inducing (PI) Wolbachia bacteria cause parthenogenesis in hosts (or thelytoky in the case of species that are already arrhenotokous) by manipulating chromosome behaviour in unfertilized eggs (Stouthamer & Kazmer, 1994). Werren (1998) stated that “by causing the rapid development of parthenogenesis within populations, PI Wolbachia may promote the evolution of parthenogenetic “species”“. The concept is that, given sufficient time, the parthenogenetic “species” accumulate deleterious mutations in the genes involved in sexual reproduction and consequently fail to fertilize their eggs (Werren, 1998). The infectious parthenogenesis is clearly a special case of sympatric speciation for which geography hardly matters. Indeed, it is difficult to imagine an allopatric origin for a parthenogenetic lineage. Wiley (1981) focused on speciation by apomixis as one example of sympatric speciation and stated that “asexual species derived from a biparental ancestral species may arise in place and there is no a priori reason to believe that such speciation must involve geographic barriers because “reproductive isolation” is instantaneous”. However, speciation caused by PI bacteria in nature has never been reported.

The common form of reproduction in Hymenoptera is arrhenotoky; males are produced from unfertilized, haploid eggs and females from fertilized, diploid eggs. Thelytokous parthenogenesis, in which females produce daughters without mating, is also found in the Hymenoptera (Heimpel & de Boer, 2008). Thelytoky may be coded for by nuclear genes of the wasp itself or by its endosymbionts such as Wolbachia (Stouthamer et al., 1993), Cardinium (Zchori-Fein et al., 2001) and Rickettsia (Hagimori et al., 2006). In this article, we will use arrhenotoky to refer to sexual reproduction and thelytoky to refer to asexual reproduction.

The solitary endoparasitoid wasp, Neochrysocharis formosa (Westwood) (Hymenotera: Eulophidae), is one of the most important natural enemies of the American serpentine leafminer, Liriomyza trifolii (Burgess) (Diptera: Agromyzidae), and the vegetable leafminer, L. sativae Blanchard, in Japan (Saito et al., 1996; Tokumaru & Abe, 2006). Thelytokous reproduction in N. formosa is known to be induced by a Rickettsia bacterium (Hagimori et al., 2006), and the cytological mechanism of diploidization by the Rickettsia is functionally apomictic (Adachi-Hagimori et al., 2008). Arakaki & Kinjo (1998) found highly female-biased sex ratios in the field, and based on the completely thelytokous laboratory strain, they suggested that in the field, both arrhenotokous and thelytokous forms occur sympatrically. However, whether gene flow occurs between the two strains has never been investigated. In addition, few data are available on the prevalence of the wasp in the field.

The purpose of the present study was to determine whether the arrhenotokous and thelytokous strains of N. formosa were the result of sympatric speciation induced by infection with a symbiotic bacterium, Rickettsia. To this end, we first collected the wasps throughout Japan and conducted PCR to see whether arrhenotokous and thelytokous strains were sister groups and whether arrhenotokous and thelytokous strains occurred sympatrically. Next, we investigated whether paternal genes from one strain could be transmitted to the other strain to determine whether the two strains are reproductively isolated from each other.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Occurrence of arrhenotokous and thelytokous strains of N. formosa in Japan

Collection and preservation of the N. formosa samples

Eggs and larvae of N. formosa that parasitized larvae of Liriomyza spp. or Chromatomyia horticola (Diptera: Agromyzidae) were collected throughout Japan from 2004 to 2007. In each field, the parasitized larvae of Liriomyza spp. or C. horticola were collected within a 5-m range so as not to collect different wasp populations. The dead bodies of Liriomyza spp. and C. horticola were incubated at 24 °C until adult wasp emergence. Details of the sites and hosts of collection are shown in Fig. 1 and Table 1.

image

Figure 1.  Collection sites of the arrhenotokous and thelytokous strains of Neochrysocharis formosa in Japan. Areas of pie charts represent composite reproductive modes frequency. Sample sizes are shown in the middle of each pie chart.

Download figure to PowerPoint

Table 1.   Collection details of Neochrysocharisformosa, N. okazakii and Closterocerus chamaeleon samples.
AbbreviationSpeciesPrefectureCountryHost flyReproductive modeCollection time (month/year)
  1. NF, Neochrysocharis formosa; NO, N. okazakii; CC, Closterocerus chamaeleon; L, Liriomyza spp., C, Chromatomyia horticola; AR, Arrhenotoky; TH, Thelytoky.

AR-SHINFShizuokaJapanCARMay/2007
TH-SHINFShizuokaJapanCTHMay/2007
AR-NAG1NFNaganoJapanCARMay/2007
AR-NAG2NFNaganoJapanCARMay/2007
AR-KYO1NFKyotoJapanLARJun/2005
AR-KYO2NFKyotoJapanLAROct/2005
TH-KYONFKyotoJapanLTHOct/2005
AR-HIRONFHiroshimaJapanCARApr/2007
TH-HIRO1NFHiroshimaJapanCTHApr/2005
TH-HIRO2NFHiroshimaJapanCTHApr/2005
AR-KAGANFKagawaJapanCARMay/2006
TH-KAGANFKagawaJapanCTHMay/2006
TH-TOKU1NFTokushimaJapanCTHMay/2006
TH-TOKU2NFTokushimaJapanCTHMay/2006
TH-OKI1NFOkinawaJapanLTHApr/2004
TH-OKI2NFOkinawaJapanLTHApr/2004
NONOHai PhongVietnamUnknownARSep/1998
CCCCSydneyAustraliaUnknownTHOct/2006
PCR amplification of nuclear, mitochondrial and endobacterial genes of N. formosa

The samples listed in Table 1 were individually homogenized in 30 μL of a Tris–EDTA buffer (5 N NaCl, 0.5 mm EDTA [pH 8.0] and 1 m Tris–HCl [pH 8.0]) and incubated with proteinase K (0.5 mg ml−1) at 37 °C for 0.5 h. The homogenates were boiled at 99.9 °C for three min to inactivate the proteinase K and used as templates for the PCR.

To amplify the nuclear gene of N. formosa, N. okazakii (Kamijo) and Closterocerus chamaeleon Girault, the primers 28SF3633 (5′-TAC CGT GAG GGA AAG TTG AAA-3′) and 28SR4076 (5′-AGA CTC CTT GGT CCG TGT TT-3′) (Tiawsirisup et al., 2008) that begin at stem 8 of the 28S ribosomal RNA sequence, encompassing the D2 loop, and end in stem 13 were used for PCR in 33-μl reactions containing 1.5 U of Taq polymerase (Applied Biosystems, Foster City, CA, USA), 0.66 μL of 10 mm dNTPs, 1.3 μL each of 10 pmol μl−1 forward and reverse primers, 3.3 μl of 10× PCR buffer with MgCl2, 1 μL of DNA template and 26.2 μL of sterile water. PCR amplifications were carried out in an ABI thermal cycler (PE Applied Biosystems PCR System 9700; PE Applied Biosystems) with the following programme: an initial denaturing step at 94 °C for 2 min; 38 cycles of 94 °C for 30 s, an annealing step at 58 °C for 50 s, 72 °C for 1 min 30 s; and a final extension step of 72 °C for 10 min.

To amplify the mitochondrial gene of N. formosa and C. chamaeleon, the primers known to amplify the mitochondrial cytochrome oxidase-I (COI) gene, COI 1 (5′-CTT TAT CAA CAT TTA TTT TGA TTT TTT-3′) and COI 2 (5′-TAC TCC AAT AAA TAT TAT AAT AAA TTG-3′) (Hoshizaki & Shimada, 1995), were used for PCR as described by Hagimori et al. (2006).

To check for bacterial infection in the samples, the bacterial general primers 16S27f and 16S1495r (Weisburg et al., 1991) were used as described by Hagimori et al. (2006). The samples that showed bacterial infection by the PCR were subsequently used for PCRs by Wolbachia-, Cardinium- and Rickettsia-specific primers.

The presence of Wolbachia was checked using Wolbachia 16Sr DNA-specific primers, V1 (5′-TTG TAG CCT GCT ATG GTA TAA CT-3′) and V6 (5′-GAA TAG GTA TGA TTT TCA TGT-3′) (O’Neill et al., 1992) with the following programme: an initial denaturing step at 95 °C for 1 min; 30 cycles of 95 °C for 1 min, an annealing step at 52 °C for 1 min, 72 °C for 1 min; and a final extension step of 72 °C for 1 min 30 s. The Wolbachia-infected wasp, Trichogramma deion Pinto and Oatman, was used as a positive control for the PCR.

The presence of Cardinium was checked using Cardinium 16Sr DNA-specific primers, EPS-F (5′-TAC AAT CTT TAT TAA CCC ATG TT-3′) and EPS-R (5′-TTC AAA GTA GCA AAA TAC ATT C-3′) (Zchori-Fein et al., 2001) with the following programme: an initial denaturing step at 95 °C for 2 min; 40 cycles of 92 °C for 30 s, an annealing step at 50 °C for 30 s, 72 °C for 30 s; and a final extension step of 72 °C for 5 min. The Cardinium-infected mite, Brevipalpus californicus Banks, was used as a positive control for the PCR.

The presence of Rickettsia was checked using a bacteria general forward primer 16S27f (Weisburg et al., 1991) and a Rickettsia-specific reverse primer Rick16SR (5′-CAT CCA TCA GCG ATA AAT CTT TC-3′) (Fukatsu et al., 2000) with the following programme: an initial denaturing step at 92 °C for 1 min; 35 cycles of 92 °C for 1 min, an annealing step at 50 °C for 1 min, 72 °C for 1 min 30 s; and a final extension step of 72 °C for 1 min 30 s.

All PCRs included a negative control (sterile water instead of DNA) to spot any DNA contamination.

Successful amplification was determined by electrophoresing 3 μL of the PCR mixture on a 2% agarose gel (1 × TAE), staining with ethidium bromide, and observing under an UV transilluminator.

DNA sequencing and BLAST search

All of the 28Sr DNA, COI and Rickettsia PCR products were sequenced directly as described in Hagimori et al. (2006).

All of the Rickettsia partial sequences were edited and assembled with the Contig Express program of the Vector NTI Advance ver. 10.1 (Invitrogen InforMax, Frederick MD, USA). Multiple alignments of the sequences were performed by the program package Clustal W (Thompson et al., 1994). The Rickettsia 16S rRNA sequence was compared with other known Rickettsia sequences by using an advanced BLAST (National Center for Biotechnology Information) search.

Phylogenetic analysis

All of the 28Sr DNA and COI sequences were edited and assembled with the Contig Express program of the Vector NTI Advance ver. 10.1. Multiple alignments of the sequences were performed by the program package Clustal W (Thompson et al., 1994). To analyse the phylogenic relationships among the wasps examined, we constructed phylogenetic trees by the neighbour-joining (NJ) method (Saitou & Nei, 1987) using Kimura’s two-parameter correction (Kimura, 1980), which is incorporated into MEGA4 program (Tamura et al., 2007). The robustness of the results was evaluated based on 1000 bootstrap replicates. The homologous sequences of C. chamaeleon for both genes were used as out groups.

Gene flow between arrhenotokous and thelytokous strains in the laboratory

Insect cultures

For crossing experiments, the thelytokous strain of the wasp was established from a stock culture originally from the Western districts of Japan at Sumitomo Chemical Co. Ltd. The arrhenotokous strain of the wasp was collected from Shizuoka, Japan. These two strains were subsequently maintained as described in Hagimori et al. (2006).

Antibiotic treatments

To obtain males from the thelytokous strain, a mixture of antibiotic tetracycline hydrochloride with honey (50 mg ml−1) was fed to newly emerged thelytokous females. Resulting male progenies were used for crossing experiments.

Sperm production

To detect sperm production, the reproductive tracts of thelytokous and arrhenotokous males (n = 10, each) were dissected, squashed and examined in a drop of 0.9% NaCl solution on a microscope slide under a stereoscopic microscope.

Sperm transfer from thelytokous males to arrhenotokous females

To test whether thelytokous males exhibit courtship behaviour to arrhenotokous females, thirteen couples of thelytokous males and arrhenotokous females were used. Each couple was confined in a test tube (5 mm diameter × 20 mm length), and male’s courtship behaviour (wing fanning) was observed under a microscope for 15 min. As a control, eighteen couples of arrhenotokous males and arrhenotokous females were confined and observed in the same manner.

To test whether arrhenotokous females use the sperm of thelytokous males to produce fertilized offspring, twelve couples of thelytokous males and arrhenotokous females (=treatment group) were used. As a control, thirteen couples of arrhenotokous males and arrhenotokous females (=control group) were used. In each cross, an unmated female was enclosed with a male in a glass vial sealed with parafilm and provided with honey for 48 h. Then, the females were allowed to oviposit in approximately 30 L. sativae larvae in a leaf of the kidney bean Phaseolus vulgaris L. for a period of 24 h. After the first oviposition period, the females were transferred to a new set of 30 host larvae and allowed to oviposit for another 24 h. This was repeated for a third time. The larvae containing wasp eggs were incubated at 24 °C, 75% RH and 16L-8D conditions. Numbers of emerging males and females were recorded for a period of 2 weeks, during which time most parasitoids emerged. For the statistical analysis, t-tests were conducted to find significant differences between the treatment group and the control group at the 0.05 alpha level. The statistical analysis of the percentage of males was conducted following an arcsine transformation (arcsine inline image, where F is the number of female offspring and M is the number of male offspring).

Sperm transfer from arrhenotokous males to thelytokous females

To test whether thelytokous females use the sperm of arrhenotokous males to produce fertilized offspring, eleven couples of arrhenotokous males and thelytokous females were tested under the same conditions as described for the previous experiment. The parents and their F1 offspring produced were collected and stored at −80 °C for the genetic analysis. The F1 offspring were checked for the presence of paternal genes using the multiplex method by Adachi-Hagimori & Miura (2008). Briefly, if the female offspring was fertilized, both paternal and maternal bands would be seen. However, if the female offspring was produced thelytokously, only maternal bands should appear.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Occurrence of arrhenotokous and thelytokous strains of N. formosa in Japan

Figure 1 and Table 1 show the occurrence of arrhenotokous and thelytokous strains of N. formosa in Japan. Both strains occur sympatrically in Shizuoka, Kyoto, Hiroshima and Kagawa Prefectures.

PCR amplification of nuclear gene, mitochondrial gene and endobacterial gene of N. formosa

The 28S rRNA gene and COI gene were successfully amplified by PCR. All sequences have been deposited in the EMBL, GenBank and DDBJ nucleotide sequence databases under the accession numbers listed in Table 2. All of the thelytokous individuals were only infected with Rickettsia (Table 2). The sequences have also been deposited in the EMBL, GenBank and DDBJ nucleotide sequence databases. There was no variation among 142 bp of the Rickettsia sequences. The Rickettsia bacterium was identical with the Rickettsia endosymbiont of N. formosa previously collected from Shizuoka, Japan (142/142 bp identity) [DDBJ Accession no. AB231472; Tagami et al. (2006)]. None of the arrhenotokous individuals were infected by bacteria.

Table 2.   28S rRNA sequence, COI sequence and bacteria infection of Neochrysocharis formosa, N. okazakii and Closterocerus chamaeleon.
AbbreviationSpeciesReproductive mode28S rDNACOIBacteria infection
  1. NF, Neochrysocharis formosa; NO, N. okazakii; CC, Closterocerus chamaeleon; AR, Arrhenotoky; TH, Thelytoky; R, Infected with Rickettsia; U, Uninfected with any bacteria.

AR-SHINFARAB508820AB509263U
TH-SHINFTHAB508821AB509264R (AB510360)
AR-NAG1NFARAB508822AB509265U
AR-NAG2NFARAB508823AB509266U
AR-KYO1NFARAB508824AB509267U
AR-KYO2NFARAB508825AB509268U
TH-KYONFTHAB508826AB509269R (AB510361)
AR-HIRONFARAB508829AB509270U
TH-HIRO1NFTHAB508827AB509271R (AB510362)
TH-HIRO2NFTHAB508828AB509272R (AB510363)
AR-KAGANFARAB508831AB509273U
TH-KAGANFTHAB508830AB509274R (AB510364)
TH-TOKU1NFTHAB508832AB509275R (AB510365)
TH-TOKU2NFTHAB508833AB509276R (AB510366)
TH-OKI1NFTHAB508834AB509277R (AB510367)
TH-OKI2NFTHAB508835AB509278R (AB510368)
NONOARAB526861N/AU
CCCCTHAB508836AB509279U
Phylogenetic analysis

The phylogenetic analysis of the 28S rRNA gene revealed that the arrhenotokous and thelytokous strains fell into two distinct clades (Fig. 2), consistent with the previous report by Adachi-Hagimori & Miura (2008). Moreover, phylogenetic analysis of cytoplasmic gene showed that all arrhenotokous strains fell into one clade, whereas all thelytokous strains fell into another clade (Fig. 3). Thus, both in the nuclear level and in mitochondrial level, the two reproductive modes were completely dissimilar (Figs 2 and 3).

image

Figure 2.  A nuclear 28S rRNA phylogeny of the arrhenotokous and thelytokous strains of Neochrysocharis formosa, N. okazakii and Closterocerus chamaeleon (outgroup). The tree was constructed by the neighbour-joining method using Kimura’s two-parameter correction. The numbers above branches indicate a bootstrap value of >70% of 1000 replicates. The bars indicate 0.01 substitutions per site. A total of 447 positions in the final dataset.

Download figure to PowerPoint

image

Figure 3.  A mitochondrial COI phylogeny of the arrhenotokous and thelytokous strains of Neochrysocharis formosa and Closterocerus chamaeleon (outgroup). The tree was constructed by the neighbor-joining method using Kimura’s two-parameter correction. The numbers above branches indicate a bootstrap value of >70% of 1000 replicates. The bars indicate 0.01 substitutions per site. A total of 443 positions in the final dataset.

Download figure to PowerPoint

Gene flow between arrhenotokous and thelytokous strains in the laboratory

Sperm production

All of the thelytokous and arrhenotokous males carried sperm in their reproductive tracts.

Sperm transfer from thelytokous males to arrhenotokous females

None of thelytokous males exhibited courtship behaviour (wing fanning) to arrhenotokous females, whereas all arrhenotokous males showed courtship behaviour to arrhenotokous females.

The sex ratio (proportion of male offspring) produced by arrhenotokous females confined with thelytokous males (100%) was significantly different from those produced by arrhenotokous females that were confined with arrhenotokous males (Table 3). Thus, thelytokous males did not fertilize the eggs of arrhenotokous females. The mean numbers of offspring produced by these females were not significantly different.

Table 3.   Crossing experiment between arrhenotokous (AR) and thelytokous(TH) strains of Neochrysocharis formosa.
Crossing combinationNo of pairs testedMean no of offspring*Mean % of males†
  1. A statistical analysis of the mean percentage of males was conducted after arcsine transformation.

  2. *Mean (±SE) numbers of offspring were not significantly different at the 5% level by t-test.

  3. †Mean (±SE) percentage of males was significantly different at the 5% level by t-test (P < 0.001).

AR♀ × AR♂1320.9 ± 3.345.0 ± 6.8a
AR♀ × TH♂1220.3 ± 3.4100.0 ± 0.0b
Sperm transfer from arrhenotokous males to thelytokous females

Arrhenotokous paternal markers were not incorporated into the genome of the 63 offspring in any of the eleven couples analysed (Table 4). Hence, arrhenotokous males did not fertilize the eggs of thelytokous females.

Table 4.   Sperm utilization by Neochrysocharis formosa thelytokous females when confined with arrhenotokous males in a glass tube.
No. of parents testedNo. of offspring testedNo. of unique loci
MaternalPaternal
1163630

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The results presented in this study demonstrate that the arrhenotokous and thelytokous strains of N. formosa underwent sympatric speciation through being infected by a symbiotic bacterium Rickettsia. First, Rickettsia infection was detected in all sampled individuals of the thelytokous strain and no individuals of the arrhenotokous strain, and there were no nucleotide differences among individuals of the thelytokous strain, which indicated that the origin of the infection occurred only once. Second, the arrhenotokous and thelytokous strains were collected in areas of sympatry (Fig. 1 and Table 1). Together with the collection data by Arakaki & Kinjo (1998) in the southern part of Japan, the two strains of N. formosa occur sympatrically in almost all parts of Japan. Third, the arrhenotokous and thelytokous strains are genetically isolated from each other at both the nuclear 28S rRNA gene and mitochondrial COI gene suggesting that gene flow does not occur between them in the field (Figs 2 and 3). The genetic relationship between the two strains has already been investigated by Adachi-Hagimori & Miura (2008). They have shown that the two strains have distinct genetic background at the nuclear first internal transcribed spacer (ITS-1) region. However, the phylogenetic status of cytoplasmic genes is sometimes different from that of nuclear genes (Sota & Vogler, 2001). Thus, the present study is important because independent phylogenetic trees were constructed using both nuclear and cytoplasmic genes. Additionally, no gene flow was detected between the two strains by crossing experiments in laboratory studies (Tables 3 and 4). This result is important because few studies have conducted crossing experiments to demonstrate reproductive isolation. Fourth, the arrhenotokous and thelytokous strains of N. formosa showed a sister relationship in the nuclear 28S rRNA gene (Fig. 2). This result is consistent with the morphological analysis of the two strains in which they are also most closely related when compared with N. okazakii as the outgroup (K. Kamijo, pers. comm.). Finally, a strong reason is presented to support that only sympatric speciation could account for the origin of the thelytokous strains of N. formosa. The thelytokous strains collected throughout Japan were only infected with Rickettsia (Table 2). Thelytokous females treated with antibiotics produced Rickettsia-free male offspring of the same reproductive form as that of arrhenotokous females (Hagimori et al., 2006). Thus, the thelytokous strain should be the result of rapid sympatric speciation from an arrhenotokous strain.

There was 1.3% nucleotide difference (462/468 bp identity) in the 28S rRNA gene and about 8% nucleotide difference in the COI gene between arrhenotokous and thelytokous strains. In Eulophidae, the 28S rRNA gene is known to be most informative at the level of subfamily and tribe (Gauthier et al., 2000), which indicates that the gene is a conservative region that evolves relatively slow. Thus, the divergence time of the two strains should be relatively old. Based on the commonly used mitochondrial DNA (mtDNA) clock rate of 2.3% pairwise divergence/million years ago (Brower, 1994), the thelytokous strain diverged from arrhenotokous strain at least 3.5 million years ago.

How did reproductive isolation occur between arrhenotokous and thelytokous strains of N. formosa? The following scenario may have occurred: (I) An individual of the arrhenotokous strain was infected with parthenogenesis-inducing Rickettsia. (II) The Rickettsia-infected female started to reproduce thelytokously. (III) Because the Rickettsia-infected female reproduces apomictically (ameiotic) (Adachi-Hagimori et al., 2008), no sperm was able to result in fertilization even though the females mated with males. (IV) From the time of infection, the arrhenotokous and thelytokous strains have accumulated different mutations in genes, resulting in different evolutionary units over time. Examples of such genetic differences could include those involved in male sexual functions such as spermatogenesis and sperm performance, and female sexual functions such as pheromone production and mate acceptance (Werren, 1998). Zchori-Fein et al. (1992) reported that males of the Wolbachia-infected wasp, Encarsia formosa Gahan, failed to inseminate asexual or cured conspecific females. Although males of the thelytokous strain of N. formosa produced sperm, they did not fertilize eggs of arrhenotokous strains likely because the thelytokous males exhibited no courtship behaviour.

In many Trichogramma species, both arrhenotokous and thelytokous strains are maintained (Stouthamer & Kazmer, 1994). A mathematical model has shown that the primary mechanism maintaining low-level PI Wolbachia infection in Trichogramma populations is reduction in survivorship from egg to adult in infected relative to uninfected females (Miura et al., 2010). Although sympatric speciation in N. formosa is suggested in this article, it is still not clear how the arrhenotokous and thelytokous strains are maintained sympatrically. The issue would solve the paradox of sex: the persistence of sexual reproduction despite its overwhelming costs.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

We thank Professor R. Stouthamer for critical reading of the original manuscript and two anonymous reviewers for useful comments. Thanks are also due to the Deciding Editor, Dr Rhonda R. Snook, for extensive and useful suggestions. We wish to thank M. Adachi, J. Imai and farmers for helping us to collect N. formosa strains. The authors also thank Dr. J. LaSalle and N. Fisher for providing wasp specimens. This research is partially supported by Grant-in-Aid for JSPS Fellows to T. A-H. Y. A. is grateful for support from the Global COE Program (Center of excellence for Asian conservation ecology as a basis of human-nature mutualism), MEXT, Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References