1 Introduction
Natural enemies of the family Trichogrammatidae are released worldwide against a number of lepidopterous pest on corn, rice, sugar-cane, cotton, vegetables and pines (Smith, 1994; Li, 1994; Grenier, 1994). In the past, most successful biological control agents have been found through trial and error. Nowadays, there is an effort to predict the success of a Trichogramma strain before introductions are made (van Lenteren and Woets, 1988).
Therefore, standard procedures to compare Trichogramma candidates for biological control have been proposed (Hassan, 1994). According to Hassan (1995) before field releases are undertaken, a suitable Trichogramma strain of known qualities should be chosen. The effectiveness of Trichogramma in the field largely depends on its searching behaviour, host preference and tolerance to environmental conditions. However, correct identification of the Trichogramma species that is going to be tested, is a crucial step in any biological control program.
Despite the large improvement with the discovery of male genitalic morphology as species specific characters (Nagarkatti and Nagaraja, 1968, 1971), the identification of these tiny insects remains difficult and requires specialized skills (Pinto and Stouthamer, 1994).
Their small size and lack of morphologically distinct characters has historically been a problem in taxonomic studies. Specimen must be prepared on slides for examination, a time consuming process that requires considerable experience (Platner et al., 1999).
Unfortunately many important species share similar genitalic structures and this has forced workers to continue relying on less dependable characters that often are intraspecifically variable and subject to phenotypic plasticity (Pinto et al., 1989; Pinto and Stouthamer, 1994).
To simplify the Trichogramma species identification, several methods have been proposed: (1) Allozymic analysis (Pintureau and Babault, 1980, 1981, 1982; Pintureau and Keita, 1989; Kazmer, 1991; Pinto et al., 1992, 1993; Pintureau, 1993). Esterase electrophoresis has provided consistent results for differentiation of some Trichogramma species (Pinto et al., 1992, 1993; Silva et al., 1999). The limited variation at esterase loci only allows for the differentiation between a limited number of species (Richardson et al., 1986; Pinto and Stouthamer, 1994). In addition the samples subjected to allozyme electrophoresis have to be kept at –70 °C to avoid the degeneration of the enzymes; (2) Reproductive compatibility tests (Nagarkatti and Nagaraja, 1968; Pinto et al., 1991; Pintureau, 1991) in which individuals of the unknown species are crossed with individuals of known species status; (3) The DNA sequence of ribosomal spacers (Landry et al., 1993; Orrego and Agudelo-Silva, 1993; Sappal et al., 1995; van Kan et al., 1996, 1997; Pinto et al., 1997); (4) RFLP’s of complete mitochondrial genome has also been used for species differentiation (Vanlerberghe-Masutti, 1994). Ribosomal DNA is present in all organisms and is composed of several regions (genes and spacers) that evolve at different rates (Hillis and Dixon, 1991); (5) The internal transcribed spacer regions (ITS-1 and ITS-2) have been used at species and intraspecific levels of many taxa for distinguishing the different forms (Carbone and Kohn, 1993; Bowles and McManus, 1993; Hsiao et al., 1994; Buckler IV et al., 1997). In insects these sequences have been often used for taxonomic purposes (Campbell et al., 1993; Hoy, 1994; Kuperus and Chapco, 1994; Vogler and DeSalle, 1994; Stouthamer et al., 1999; Silva et al., 1999; Chang et al., 2001; Pinto et al., 2002; Honda et al., 2006; Dem'yanchuk et al., 2008; Kumar et al., 2009; Polaszek et al., 2011; Poorjavad et al., 2012).
Here, we use the ITS2 sequences of Trichogramma species (native or introduced) from South America to develop an identification key. The advantage of this identification system over the morphology-based system is that non-specialists are able to quickly and cheaply identify individual specimens. In addition females can also be identified which is not possible in the morphologically based system. Species found in South America are thelytokous, either because of infection with parthenogenesis inducing Wolbachia (T. pretiosum and T. atopovirilia) (Grenier et al., 1998; Almeida et al., 2001, 2010; Ciociola Junior et al., 2001a) or because of some nuclear genetic factor (T. cacoeciae Marchal) (Stouthamer et al., 1990; Almeida and Stouthamer, 2003). Variation of ITS2 sequence within Trichogramma species is relatively small in comparison to the difference found between species and all morphologically distinct cryptic species are also distinguished by sequence differences (Stouthamer et al., 1999). ITS2 sequences within each species are very similar and there is no evidence for two or more gene families that differ substantially within the genome of a single individual, as has been found in other taxa (Vogler and DeSalle, 1994).
2 Material and Methods
2.1 Trichogramma cultures
Table 1 lists the Trichogramma species/lines studied here, including their origin, host insect, plant crop, collection date and status as native or introduced.
Table 1 Trichogramma populations from South America.
Spécies | Line Designation | Origin | Host Insect | Plant Crop | Collection Date |
---|---|---|---|---|---|
T. acacioi Brun, Moraes and Soares, 1984 | Taca-01 | Botucatu-SP-Brazil | Euselasia sp. | - | - |
T. atopovirilia Oatman and Platner, 1983 | Tato-01 | Sete Lagoas-MG-Brazil | - | Corn | - |
Tato-02 | Colombia | - | - | - | |
T. brassicae Bezdenko, 1968 | Tbra-01 | Introduced from Germany | - | - | Dec., 1996 |
T. bruni Nagaraja, 1983 | Tbru-01 | Piracicaba-SP-Brazil | Heliconius phyllis | Passiflora sp. | - |
T. cacoeciae Marchal, 1927 | Tcac-01 | Huarochiri-Peru | Cydia pomonela | Apple | Oct., 1997 |
T. dendrolimi Matsumura, 1926 | Tden-01 | Introduced from Germany | - | - | Dec., 1996 |
T. esalqueanumQuerino and Zucchi, 2003 | Tesa-11 | Piracicaba-SP-Brazil | Helicornius erato phyllis | Passiflora sp. | Oct., 1999 |
T. exiguum Pinto and Platner, 1983 | Texi-01 | Peru (different places) | Diatraea saccharalis | Sugar-cane, | - |
Helicoverpa zea | Corn | - | |||
Erynis ello | Cassava | - | |||
P. persimilis | Olive | - | |||
Dione juno | Passiflora P. | Mar., 2001* | |||
Texi-02 | Colombia | - | - | - | |
T. fuentesi Torre, 1980 | Tfue-01 | Peru | Diatraea saccharalis | Sugar-cane, rice | - |
Helicoverpa zea | Corn | - | |||
Heliothis virescens, Anomis texana | Cotton | Dec., 1994* | |||
T. galloi Zucchi, 1988 | Tgal-01 | Chiclayo-Peru | Diatraea saccharalis | Sugar-cane | Apr., 2000 |
T. iracildaeQuerino and Zucchi, 2003 | Tira-01 | Maceió-AL-Brazil | Calpodes ethlius | Canna spp. | Apr., 2000 |
T. lasallei Pinto, 1998 | Tlas-01 | Lima-Peru | Quinta cannae | Canna indica | Mar., 2001 |
T. lopezandinesis Sarmiento, 1983 | Tlop-01 | Colombia | - | - | - |
T. nerudai Pintureau and Gerding, 1999 | Tner-01 | Chile | Rhyacionia buolina | Pine forest | - |
T. pintoi Voegelé, 1982 | Tpin-01 | Introduction from Chile | - | - | Jun., 1973 |
T. pretiosum Riley, 1879 | Tpre-01 | Chincha-Peru | Diaphania nitidalis | Pumpkins | Mar., 1992 |
Tpre-02 | Chancay-Peru | Dione juno | Passiflora | May, 1994 | |
Tpre-03 | Lambayeque-Peru | Diatraea saccharalis | Corn | Jul., 1994 | |
Tpre-04 | Tacna-Peru | Palpita persimilis | Olive | Apr., 1995 | |
Tpre-05 | Moquengua-Peru | Marasmia trapezalis | Corn | Jul., 1996 | |
Tpre-06 | Ica-Peru | Heliothis virescens | Cotton | Nov., 1996 | |
Tpre-07 | Piura-Peru | Alabama argillacea | Cotton | Apr., 1997 | |
Tpre-08 | Lima-Peru | Diaphania nitidalis | Curcubita moschata | Mar., 1997 | |
Tpre-09 | Ayacucho-Peru | Helicoverpa zea | Corn | Nov., 1998 | |
Tpre-10 | Colombia | Neoleucinodes elegantalis | - | - | |
Tpre-11 | USA | - | - | - | |
Tpre-12 | Jataizinho-PR-Brazil | - | Cotton | Feb., 2000 | |
T. rojasi Nagaraja and Nagarkatti, 1973 | Troj-01 | Curitiba-PR-Brazil | Anticarsia gemmatalis | - | - |
*Last collection date.
2.2 DNA Extraction, PCR Amplification and electrophoresis
To extract DNA from the different species/lines we used five wasps that were homogenized in 100 µL 5% Chelex-100 and 4 µL proteinase K (20 mg/mL) and incubated for at least 4 hours at 56 °C, followed by 10 min. at 95 °C. The PCR was performed in a total volume of 50 µL using a Techne thermocycler, 5 µL DNA template, 5 µL 10x PCR-buffer, 1 µL dNTP’s (each in a 10 mM concentration), 1 µL forward and reverse primers (ITS2-forward: 5’ -TGTGAACTGCAG GACACATG-3’ located in the 5.8S region of the rDNA; ITS2-reverse: 5’ -GTCTTGCC TGCTCTGCTCTGAG-3’ located in the 28S region of the rDNA (Stouthamer et al., 1999); 0.14 µL TAQ polymerase (5 units/µL) and 36.86 µL of sterile distilled water. The cycling program was 3 min. at 94 °C followed by 33 cycles of 40 seconds at 94 °C, 45 second at 53 °C and 45 seconds at 72 °C with 5 min. at 72 °C after the last cycle.
2.3 Cloning, sequencing and alignments
Following electrophoresis, PCR products were purified with a QIAquick PCR purification kit (Qiagen®). After the purification the PCR products were ligated into a Pgem-T®Vector (Promega), 2 µL of the ligation mix was transformed in the heat shock cells of DH5-α Escherichia coli and plated on a LB agar medium containing Ampicilin, XGAL and IPTG. The plates were incubated overnight at 37 °C. The next day, white colonies were picked up with a sterile toothpick from the plates and placed into tubes containing 3.0 mL of LB liquid medium and 3 µL Ampicilin and put to grow up overnight in a shaker set to 250 rpm at 37 °C. To confirm that the correct piece of DNA had been cloned, a PCR reaction with a template extracted from the bacterial culture was added to 100 µL 5%
Chelex-100 and incubated for 15 min. at 60 °C followed by 5 min. at 95 °C. The PCR was performed with 5 µL of this template as described before. If indeed an ITS2 had been cloned, 850 µL of the bacteria culture was added to 150 µL of 87% glycerol and stored at –80 °C. The rest of the culture was used in a QIAprep Miniprep kit (Qiagen®) to purify the plasmid, which was used for the sequencing in an Applied Biosystems automatic sequencer. Trichogramma sequences were aligned manually using the ESEE 3.0s sequence editor (Cabot, 1995).
2.4 Thelytoky in Trichogramma
For detection of thelytoky caused by Wolbachia infection, specific primers for DNA amplification of the wsp region of Wolbachia were used: wsp-Forward primer 5’TGGTCCAATAAGTGATGAAGAAAC-3’ and wsp-Reverse 5’ –AAAAATTAAACGCTACTC CA-3’ (Braig et al., 1998). These primers amplify 554bp of the wsp gene. The cycling program was 3 min. at 94 °C followed by 40 cycles of 1 min. at 94 °C, 1 min. at 50 °C and 1 min. at 72 °C with 5 min. at 72 °C after the last cycle.
2.5 Molecular key
A molecular key for separating Trichogramma species was constructed based on the size and the fragment profile following restriction with different restriction enzymes. Initially we determined ITS2 sequences of all species and lines, subsequently we first tried to distinguish the different species based on the size of the PCR product. Species with similar sized PCR products were distinguished by selecting restriction enzymes that would generate differently sized in restriction fragments. To predict the fragment sizes to be expected with different enzymes we use the program Webcutter 2.0 (Heiman, 1997). Three restriction enzymes were used (EcoRI, MseI and MaeI).
3 Results
Seventeen native and introduced Trichogramma species collected in South America were identified based on ITS2 sequences. In total, thirty lines were studied (Table 1). Seventeen lines were recognised as arrhenotokous and thirteen as thelytokous. Twelve out of thirteen thelytokous lines studied here carry the Wolbachia symbiont. In T. cacoeciae, parthenogenesis was not caused by Wolbachia infection (Stouthamer et al., 1990; Almeida and Stouthamer, 2003). For all other parthenogenetic lines, Wolbachia detection was possible by using specific primers of the wsp region. In all thelytokous species/lines no male was found, except for T. atopovirilia (culture form Brazil) and T. pretiosum (culture from the Entomology Lab., California University, Riverside-USA) in which few males were present.
Complete ITS2 sequences have been deposited in GenBank (Table 2). Trichogramma species were identified by (1) morphological identification: T. atopovirilia, line Tato-01, (Dr. Américo I. Ciociola Junior - EPAMIG, Brazil, personal comunication), T. lopezandinensis and T. fuentesi (Dr. Ranyse B.Q. da Silva-ESALQ/USP, Brazil, personal communication); (2) by comparison with ITS2 sequences from the GenBank: T. pintoi, T. lasallei, T. cacoeciae, T. galloi, T. pretiosum, T. brassicae, T. dendrolimi and T. exiguum; or (3) by comparison with sequences not published yet: T. nerudai. T. rojasi and T. acacioi lines were the same as those used by Ciociola Junior et al. (2001b). Two new species from Brazil (T. iracildae and T. esalqueanum) sequenced here were recorded by Querino and Zucchi (2003).
Table 2 Line designation, reproduction mode, genbank accession number of the Trichogramma species and size of the ITS2 and PCR product (bp).
Line Designation | Reproduction Mode | Acession Number | ITS-2 Product | PCR product |
---|---|---|---|---|
Tner-01 | Arrhenotoky | AY182756 | 632 | 746 |
Tpin-01 | Arrhenotoky | AY182757 | 581 | 695 |
Tato-01 | Thelytoky | AY182758 | 565 | 680 |
Tato-02 | Arrhenotoky | AY182759 | 561 | 675 |
Taca-01 | Arrhenotoky | - | 559 | 674 |
Tbru-01 | Arrhenotoky | AY187263 | 536 | 650 |
Troj-01 | Arrhenotoky | - | 524 | 638 |
Tira-01 | Arrhenotoky | AY182760 | 522 | 636 |
Tlop-01 | Arrhenotoky | AY182761 | 509 | 623 |
Tlas-01 | Arrhenotoky | AY182762 | 485 | 599 |
Tesa-11 | Arrhenotoky | AY182763 | 379 | 493 |
Tcac-01 | Thelytoky | AY166700 | 460 | 574 |
Tgal-01 | Arrhenotoky | AY182764 | 445 | 560 |
Tfue-01 | Arrhenotoky | AY182765 | 437 | 553 |
Tpre-01 | Thelytoky | AY182770 | 410 | 524 |
Tpre-02 | Thelytoky | AY182771 | 410 | 524 |
Tpre-03 | Thelytoky | AY182772 | 415 | 529 |
Tpre-04 | Thelytoky | AY182773 | 410 | 524 |
Tpre-05 | Thelytoky | AY184958 | 410 | 524 |
Tpre-06 | Thelytoky | AY184959 | 410 | 524 |
Tpre-07 | Thelytoky | AY187259 | 415 | 529 |
Tpre-08 | Thelytoky | AY184960 | 410 | 524 |
Tpre-09 | Thelytoky | AY184961 | 410 | 524 |
Tpre-10 | Thelytoky | AY187260 | 412 | 526 |
Tpre-11 | Thelytoky | AY187261 | 413 | 527 |
Tpre-12 | Arrhenotoky | AY187262 | 412 | 526 |
Tbra-01 | Arrhenotoky | AY182766 | 406 | 520 |
Tden-01 | Arrhenotoky | AY182767 | 403 | 519 |
Texi-01 | Arrhenotoky | AY182768 | 383 | 497 |
Texi-02 | Arrhenotoky | AY182769 | 381 | 496 |
An identification key was constructed for these species using as two characters of the PCR product its size and the restriction patterns generation using the three restriction enzymes EcoRI, MseI and MaeI (Table 3). España-Luna et al. (2008) used two restricion enzymes (EcoRI and AluI) for distinguishing six Trichogramma species from each other in Mexico. Sumer et al. (2009) used three restricion enzymes (MnlI, MseI and DraI) for distinguishing six Trichogramma species from each other in agricultural settings around the Mediterranean.
Table 3 Molecular key for Trichogramma species recognition based on the size of the PCR product and species-specific banding pattern.
1. | Size of the PCR product < 620 bp | 2 |
Size of the PCR product > 620 bp | 10 | |
2. | Size of the PCR product < 550 bp | 3 |
Size of the PCR product >550 bp | 7 | |
3. | Size of the PCR product ≤ 500 bp | 4 |
Size of the PCR product > 550 bp | 5 | |
4. | PCR product cut by MseI ca. 280 and 191bp | T. esalqueanum |
PCR product not cut by MseI | T. exiguum | |
5. | PCR product cut by EcoRI | 6 |
PCR product not cut by EcoRI | T. pretiosum | |
6. | PCR product cut by MseI ca. 441 bp | T. dendrolimi |
PCR product cut by MseI ca. 411 bp | T. brassicae | |
7. | PCR product not cut by MseI | 8 |
PCR product cut by MseI | 9 | |
8. | PCR product cut by MaeI ca. 267 and 218 bp | T. fuentesi |
PCR product cut by MaeI ca. 182 and 113 bp | T. galloi | |
9. | PCR product cut by EcoRI ca. 346 and 228 bp | T. cacoeciae |
PCR product not cut by EcoRI | T. lasallei | |
10 | Size of the PCR product < 670 bp | 11 |
Size of the PCR product > 670 bp | 14 | |
11. | PCR product not cut by MseI | 12 |
PCR product cut by MseI | 13 | |
12. | PCR product cut by MaeI ca. 480 and 143 bp | T. lopezandinensis |
PCR product cut by MaeI ca. 517 bp | T. iracildae | |
13. | PCR product cut by MseI ca. 390 and 194bp | T. rojasi |
PCR product cut by MseI ca. 463 and 187bp | T. bruni | |
14. | Size of the PCR product > 700 bp | T. nerudai |
Size of the PCR product < 700 bp | 15 | |
15. | PCR product cut by EcoRI | T. pintoi |
PCR product not cut by EcoRI | 16 | |
16. | PCR product cut by MseI ca. 478 and 196 bp | T. acacioi |
PCR product cut by MseI ca. 345, 198 and 137 bp | T. atopovirilia |
Here, ITS2 product size of the seventeen studied species ranged from 379 to 632 bp and sequences differ from each other consistently. In our sample the size of the ITS2 product alone could not identify the species, but cutting the ITS2 product of the different species using only three restriction enzymes (EcoRI, MseI and MaeI) enabled us to identify all species (Table 3). For instance the variation in the size of ITS2 was evaluated in T. atopovirilia, T. exiguum and T. pretiosum. The length of the complete sequences found for T. pretiosum lines all ranged from 410 to 415 bp (Table 3), for T. atopovirilia (561-565 bp) and for T. exiguum (381-383 bp). In general, variation was limited to the number of microsatellite repeat stretches found in the ITS2 sequences.
4 Discussion
Distributed worldwide, Trichogramma are known with circa of 210 species (Pinto, 2006). In North America, Central America and South America are recorded respectively 60, 21 ad 41 species. Considering only the presence of native species in South America, Brazil has the largest number of known species (26) followed by Venezuela (13), Colombia (9), Peru (7), Uruquay (6), Chile (4) and Ecuador and Paraguay (2) (Zucchi et al., 2010; Querino and Zucchi, 2011).
In T. atopovirilia the presence of Wolbachia was recorded for the first time in Brazil by Ciociola Junior et al. (2001a). In all T. pretiosum lines studied here Wolbachia infection was reported for the first time. In South America, Wolbachia infection has been reported in T. pretiosum from Uruguay (Grenier et al., 1998) and from Brazil (Almeida et al., 2001).
The usefulness of endonucleases analysis for distinguishing Trichogramma species was shown by Stouthamer et al. (1999), Silva et al. (1999) and Pinto et al. (2002). How reliable are such molecular keys that are based in some cases on the sequence of only a few lines? The experience with other Trichogramma species is that the intraspecific variation in the ITS2 is rather limited. The variation in number of microsatellite repeats seems to be common in ITS sequences, similar microsatellite variation was for instance also found in Eriophyid mites (Fenton et al., 1997). The range obtained here confirms the low intraspecific variation in the ITS2 length detected by Stouthamer et al. (1999) and Silva et al. (1999). According to Stouthamer et al. (1999) ITS2 can be used for species identification in Trichogramma because the sequence variation within species is small relative to the difference found between species. All morphologically distinct cryptic species are also distinguished by sequence differences, however in the morphologically indistinguishable species North American species T. minutum Riley and T. platneri Nagarkatti no consistent differences were found in their ITS2 sequences (Stouthamer et al., 2000). In this case the ITS2 did not improve upon the morphologically based system. A potential weakness of an ITS2 based molecular key as presented here is that in some cases the key relies on either a restriction of an ITS2 by a particular enzyme or the lack thereof. When no restriction digestion is found it is not immediately clear if the endonuclease worked or if the products lack the restriction site (Stouthamer et al., 1999). It is important to run along a positive control to assure that the restriction reactions work. Molecular laboratories with basic infrastructure can perform PCR amplifications followed by restriction digestions and DNA can be extracted from living, frozen, ethanol-preserved or dried material (Post et al., 1993; van Kan et al., 1996; Ciociola Junior et al., 2000). Just as in case of morphological keys, molecular keys will only be completely reliable once all species of a region are known.
This study together with previous studies (Stouthamer et al., 1999; Silva et al., 1999; Ciociola Junior et al., 2001b) shows the utility of ITS2 sequences in Trichogramma identification. The molecular technique used in this study will be of a great utility in a near future for sequencing species already classified but not sequenced yet from South America. We also expect that with this technique new species will be soon discovered due to the high diversity of Trichogramma host insects in this region. The small number of species found until now is caused by the very limited activities to collect species of Trichogramma. In addition, most of the host species collected were restricted to insects of economic importance. Also there are only a few taxonomists specialized in identifying Trichogramma species using morphological features. Identification by the combination of morphological characters and molecular technique will be of extreme importance for a successful systematic analysis of this group.