Evaluation of horizontal gene transfer risk between the Mediterranean fruit fly Ceratitis capitata (Tephritidae) and its parasitoid Fopius ceratitivorus (Braconidae)

The transgenic strain of the Mediterranean fruit fly (medfly), Ceratitis capitata (Wied.) VIENNA 8 1260, developed from the classical genetic sexing strain VIENNA 8, has two molecular markers that exhibit red fluorescence in the body and green fluorescence in testicles and sperm. These traits offer a precise tool to discriminate between mass-reared sterile males and wild fertile males, and they could potentially increase the effectiveness of control programs for this pest. To assess the risk of horizontal transfer of the fluorescence transgenes in natural ecosystems, we used the VIENNA 8 1260 strain and the medfly parasitoid Fopius ceratitivorus. The fluorescence signal and the inheritance of the fluorescence gene markers were monitored for over 16 generations (about two years) in both species using fluorescence microscopy and a PCR-based assay. The PCR analysis was performed in four independent laboratories. Both fluorescence microscopy and PCR analysis indicated that no horizontal gene transfer of the DsRed transgene occurred during 16 generations of medfly parasitoid rearing under experimental conditions.


Introduction
The order Diptera, which includes mosquitoes and fruit flies, harbours several invasive insect species that can have a negative impact on agriculture and human health [1]. The distribution of invasive mosquito and fruit fly species is favored by factors such as global climate change [2], growing international trade [3] and increasingly vulnerable zones that promote pest establishment [4][5][6]. Countries and regional organizations establish control programs to eradicate them or prevent PLOS  fully sterilize insects for SIT applications ensures that there will be no vertical transmission of any transgene in nature. However, this does not prevent their potential horizontal transmission through, for example, transposable elements, symbionts, parasites or viruses [29][30][31][32]. The horizontal transfer of genes has been documented in many prokaryotic and eukaryotic species, as well as within and between kingdoms, and it's particularly pronounced in insects in respect of the transfer of transposable elements as well as genes or even entire genomes of symbiotic bacteria [33][34][35][36][37][38][39][40][41]. Although engineering the transposons used in the germ line transformation of insects minimizes their instability in transgenic insects [42], the risk of horizontal transfer of the transgenes still exists since such transfers may be mediated by other natural mechanisms. Taken together, risk assessment studies are needed to quantify, if possible, the likelihood that such an event may occur during the time frame of a typical medfly population control program, lasting from a few months up to 10 years [43]. The results of this study should only be interpreted in terms of SIT programs, but not in terms of a general evolutionary risk of HGT in insects, which has been reported in the literature and might occur over a much longer time frame [44]. Fopius ceratitivorus Wharton (Hymenoptera: Braconidae) was selected for this study, because it is a promising egg-larval parasitoid for the population control of medfly, thus providing the opportunity to monitor by means of PCR, the potential HGT of the VIENNA 8 1260 medfly fluorescence transgene to the parasitoids. This parasitoid was originally reported from fruit flies collected in coffee berries from Kenya [45], which is part of the African region where both coffee and the medfly are originated. A small colony of F. ceratitivorus has been established at the San Miguel Petapa laboratory from parasitoids collected in Kenyan coffee plantations and introduced to Guatemala through USDA-APHIS following proper quarantine and biosafety protocols [46].
Fopius ceratitivorus is currently being mass-reared in Guatemala [47] and tested in medflyinfested areas. There are ongoing efforts to mass-rear this species at Hawaii [48]. In Guatemala, coffee (Coffea arabica L.) is the main host plant for C. capitata (Wied.) and its biology, worldwide distribution, host plant preferences and genetic variability has been extensively documented [49][50][51][52][53][54][55][56][57][58]. The medfly-parasitoid interaction provides an ideal scenario to test the HGT risk between these two species, given the nature of the parasitoid's life cycle, in close contact with the medfly host. The presence of the fluorescent transgene in VIENNA 8 1260 medflies adds the possibility of detecting a potential HGT by the detection of the DsRed transgene through epifluorescence microscopy of the parasitoids, that would complement the PCR analyses and allows for a quick monitoring of HGT in adult parasitoids.

Materials and methods
In the present study, the parasitoid F. ceratitivorus was reared on the transgenic VIENNA 8 1260 strain for 16 generations, a time period representing about 2 years, with the purpose of detecting the potential horizontal transfer of the transgenes from the fly host to the parasitoid by means of fluorescent microscopy and PCR analysis.
The parasitoid Fopius ceratitivorus Wharton is routinely reared at San Miguel Petapa (SMP) laboratory, on a bisexual strain of C. capitata -Tolimán/99 as a host following a validated methodology [47]. For the purpose of the present study, the parasitoids were reared on the transgenic Ceratitis capitata strain VIENNA 8 1260 for 16 generations (G 1 to G 16 ). In each generation, VIENNA 8 1260 females laid eggs in apple (Malus domestica Borkh) fruits. These fruits were then exposed to the parasitoid F. ceratitivorus, whose adults were recovered afterwards and a new cycle started. The VIENNA 8 1260 females came from a small colony (generation 36) following the filter rearing system, FRS [59], reared at the quarantine area of the SMP laboratory, located at 15 km southeast of Guatemala city (N 142 9´2", W 90˚36´53").

Infestation of apple fruits with VIENNA 8 1260 eggs
Under controlled environmental conditions, 3000 males and 3000 females of the VIENNA 8 1260 strain were kept inside 30 × 30 × 30 cm square cages (SC) with 0.5 mesh plastic on their six sides. The flies were fed ad libitum with a mixture of 3:1 sucrose: hydrolyzed yeast and water contained in 200 ml plastic container (8.2 cm H × 6.5 cm in diameter) covered with a cotton plug. To facilitate fruit oviposition by the medfly and later its parasitoids, the apples were prepared by puncturing them with a stainless-steel pin, thus creating approximately 1500 punctures per fruit. After a mating period of 3 days, the apples (4-6 per cage) were exposed to the VIENNA 8 1260 females as an oviposition substrate. The exposure was repeated three times per cage and, in each case, two apples were left inside the cage for 24 h. All the infested apples were then kept for 24 h in a room at 24˚C and 65% relative humidity to promote the embryonic development of the VIENNA 8 1260 eggs. This procedure was repeated in each generation.

Parasitization of the VIENNA 8 1260 eggs by F. ceratitivorus
A total of 2000 males and 2000 females of the F. ceratitivorus parasitoids kept inside 30×30×30 cm 3 cubic plexiglass cages were fed with honey and water. The parasitoids were kept for 7 days, until they reached sexual maturity and mated. Then, the apples infested with VIENNA 8 1260 eggs were introduced for 24 h in the cages with the parasitoids, thus allowing the colonization of the VIENNA 8 1260 eggs by them. After the parasitization period was ended, each apple was cut into four parts and put on artificial standard larval diet to feed the parasitized VIENNA 8 1260 larvae for 6 days. This diet is a modification of the standard Seibersdorf diet described by Braga et al. (2006) [60]. The VIENNA 8 1260 larvae were collected and pupated. The VIENNA 8 1260 adults (males and females) emerged at day 12 after pupation, and the adult parasitoids (males and females) three days later. The adult parasitoids recovered in each generation were introduced again in 30×30×30 cm 3 cubic plexiglass cages and the parasitization procedure of the VIENNA 8 1260 eggs was repeated for 16 generations. In each generation, six replicates samples of 6-10 randomly selected adult parasitoids of each sex were preserved in absolute ethanol inside 1.5 ml Eppendorf vials (6 vials per sex per generation) for DNA analysis.

PCR analysis of the parasitoid samples reared on VIENNA 8 1260 flies
To evaluate the potential horizontal transfer of DNA sequences of the fluorescence transgene from the VIENNA 8 1260 flies to its parasitoid, DNA from F. ceratitivorus parasitoids reared on the transgenic flies was extracted and amplified by PCR with 1260DsRed-F (AGCTGGACAT CACCTCCCACAACG) and 1260DsRed-R (GTACTGGAACTGGGGG GACAG) (Macrogen Inc., Rockville, USA) primers specific for the fluorescence gene markers present in VIENNA 8 1260 [61]. The PCR analysis was conducted on parasitoids collected from generations G 0 , G 11 , G 12 and G 14 at the Agroecological and Environmental Biotechnology Laboratory (LaBTAA) from El Colegio de la Frontera Sur (ECOSUR), Tapachula, Chiapas, Mexico. Samples of parasitoids from generations G 0 to G 16 were also analyzed at three additional laboratories: a) Genetic Laboratory of ECOSUR, Unidad San Cristóbal (México); b) the Insect Pest Control Laboratory of the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture in Seibersdorf, Austria, and c) the Institute for Insect Biotechnology at the Justus-Liebig-University Gießen in Gießen, Germany.
DNA was extracted, individually, from all male and female parasitoids sampled from generations G 0 , G 11 , G 12 , G 14 , and G 16 . For the PCR amplification, a 1.0 μl of the extracted DNA and a 2.0 μl from a 1:10 dilution was analyzed using specific primers 1260DsRed-F and 1260DsRed-R for the detection of the DsRed fluorescence transgene. The PCR parameters were: 95˚C for 5 min; 30 cycles of 95˚C for 30 sec, 56.8˚C for 30 sec, 72˚C for 1 min; and final extension 72˚C for 5 min.
The PCR was analyzed on 1% agarose gels loaded with 10 μl out of the 20 μl of the PCR product. The size of the DNA bands was estimated by observing them under UV light and comparing their migration distance with that of the molecular weight markers coming from a 1 kb DNA ladder (O' Gene Ruler 1Kb Plus DNA Ladder, Thermo Scientific, No. Cat #SM1343). As positive control, an amplicon of 700 bp of the DsRed sequence from a VIENNA 8 1260 fluorescent adult was generated, while the negative control was DNA from a F. ceratitivorus adult parasitoid reared on a wild type strain of C. capitata, with no contact to the VIENNA 8 1260 strain.
To verify the quality of the DNA extraction and the reliability of the PCR results, another PCR was performed with a 1.0 μl subsample of the DNA previously extracted from the parasitoids samples and 2.0 μl of their 1:10 dilution using the NS1-GCFung primers [62] directed at the 18S eukaryotic ribosomal unit marker, and the following PCR conditions: 95˚C for 5 min; 35 cycles of 95˚C for 30 sec; 50˚C for 40 sec; 72˚C for 1 min; and 72˚C for 5 min. A 10 μl subsample from the 20 μl PCR product was analyzed by electrophoresis on 1% agarose gels. DNA band sizes were estimated by comparison to a 1 kb ladder molecular weight marker (O' Gene Ruler 1Kb Plus DNA Ladder, Thermo Scientific, No. Cat #SM1343). As positive control for DNA extraction, an amplicon of 500 bp of the 18S marker from a VIENNA 8-1260 fluorescent adult was generated; as the negative control was a reaction mixture with all the reagents used in the PCR processing except DNA.

Fluorescence microscopy analysis and persistence of the fluorescent protein expression through time
Complementary to the PCR analysis, a fluorescence microscopy analysis of the parasitoids reared on VIENNA 8 1260 fluorescent flies was conducted from G 0 to G 16 (1200 male and 1200 female parasitoids per generation, for a total of 40,800) in order to verify their ability to express (or not) red or green fluorescence under UV light and specific filters: GFP Plus 510/ 20-nm band pass filter for green fluorescence and Texas Red with a 75-nm emission window (608 to 683-nm) and a 55-nm excitation passband (532 to 587-nm) for red fluorescence, with the aid of a Leica MZ-FL III stereoscope.
To test the persistence of the fluorescent protein expression though time, triplicate samples of the G 0 males and females in 100 adult grids were kept under laboratory conditions (24˚C and 65% RH) with no food or water until they died (about 6 days later) and the protein expression was verified weekly under the fluorescence microscope for up to 10 weeks after their death, which is about 10 times the lifespan of a field trap. A similar triplicate set of 100 male and 100 female adults in grids was also kept under field cage conditions, with a wide range of external environmental temperature and humidity (14-33˚C, 40-100% RH) and analyzed with the same procedure.
PCR results obtained from the other three laboratories testing the same G 0 (18♂+18♀) and G 16 (18♂+18♀) samples, are summarized in Table 1. In all independent assays, no amplification of the DsRed sequence was detected in the parasitoid samples tested, while a DsRed amplicon was amplified in the positive control. The fact that all different laboratories came up with the same results strengthen the validity of the results by lowering the chance of false negative results due to failures of the PCRs. Fluorescence microscopy data are summarized in Table 1. Though this table only presents results for the same parasitoid generations analyzed by PCR, namely G 0 , G 11 , G 12 , G 14 and G 16 , in fact no fluorescence was detected in the generation by generation analysis of 1200 males and 1200 females per generation of F. ceratitivorus from G 0 to G 16 . The positive controls (VIENA 8 1260 male and female adults) showed 100% fluorescence expression for all transgenic adults parasitized at the same generations as the examined parasitoids. As an alternate way of determining a fully functional DsRed gene being horizontally transferred, the presence of the DsRed fluorescent marker in the parasitoids was checked. In an eradication program, millions or even billions of males are released in the field each week. If, for any reason (failures in the irradiation procedures, accidental releases of fertile males or intentional releases of transgenic lethality strains that have not been treated with radiation) they produce offspring, those might be colonized by the Fopius parasitoids.
Taken together, the data indicates that no horizontal gene transfer (HGT) of the DsRed fragment from the transgenic medfly host to the parasitoid genome was detected, under the experimental conditions of the study (rearing of the F. ceratitivorus on transgenic medflies for a period of 16 generations); we considered that this time period was adequate to assess the risk of HGT, taking into account that for eradications or management programs, this is a good representation.
Additionally, this study demonstrated that fluorescent protein is expressed in the VIENNA 8 1260 adults for as long as there is any soft tissues inside their exoskeletons, which was observed for at least 10 weeks after their death in all 100 male and 100 female grid samples kept under laboratory conditions. In a real world scenario, traps for monitoring are checked every week, so that this time period is about 10 times the lifespan of a field trap. The same results were obtained from field cage samples examined under the same protocol.
This study assesses the risk of horizontal transfer of the fluorescent transgene DsRed from the transgenic medfly strain VIENNA 8 1260 to F. ceratitivorus, during rearing of the parasitoid for 16 generations on its host. This time period represents about two years of repeated contact between the F. ceratitivorus parasitoids and the transgenic VIENNA 8 1260 host. Our analysis was based on fluorescence microscopy in order to see if the transgenic cassette has been transferred to the parasitoid in a way that maintained its functionality. However, no evidence of red fluorescence was detected in 8,867 parasitoids screened during sixteen generations of rearing, including males and females in a 1:1.37 ratio. In addition, we performed PCR analysis of 92 male and 92 female randomly selected parasitoids in five generations G 0 , G 11 ,  The recent use of transgenic organisms in health and agriculture has generated an intense debate between two main groups. One that opposes to the use of transgenic organisms, perceiving in their use the possibility of an horizontal transfer of genes and their potential impact in the natural ecosystems and human health, and a second one who favors their use that balance their benefits against the probability that such horizontal transfer may indeed occur. In the short term, it is such an extremely rare event, that it will unlikely play a role on the environment. However, it's well known that HGT events (a) can take place even between members of different domains of life [63] and (b) have been commonly detected in insects being mediated by transposable elements, symbionts like Wolbachia and viruses, so risk assessment analysis is needed to address the concerns of releasing transgenic insects for field applications against insect pests and disease vectors.

Conclusions
In the present study, we assessed the potential of HGT between the medfly and its parasitoid F. ceratitivorus during the typical time frame of an eradication program. We were unable to detect any HGT event. However, the data of this study should be read with caution since no detection does not necessarily mean that there have not been HGT events which we could not detect because: (a) a different part of the transgene cassette than that tried to detect by PCR was moved; (b) the transferred fragment carry mutations and this may explain the lack of fluorescence signal and PCR amplification; (c) it is practically impossible to check all individuals in every single generation and so, our PCR analysis was performed only in a limited number of insects and in just four out of sixteen generations and, last but not least, (d) nothing excludes the possibility that such an HGT event is so rare that we could not detect it in just 16 generations with that size of parasitoid population used in our experiment.
The persistence of the fluorescence expression in VIENNA 8 1260 adults for at least ten weeks under both laboratory and field cages conditions shows the stability of this fluorescent protein. Nevertheless, under different environmental conditions (higher humidity or extreme temperatures) this protein might degrade faster and further research needs to be conducted to test the durability of the protein for those conditions.