Parental RNAi Silencing of the transformer-2 Gene in a Species of the Anastrepha fraterculus Complex of Cryptic Species (Diptera, Tephritidae)

In genera Anastrepha, Bactrocera and Ceratitis of the tephritid fruit flies the auto-regulatory function of gene transformer is assumed to be activated by maternal derived mRNA or the proteins of the gene transformer (tra-2 ) and transformer-2 (tra-2). However, this maternal effect was not yet been demonstrated. The objective of the present study was to test the effect of absence of tra-2 in the eggs on the sex determination of A. sp.1 affinis fraterculus. This was achieved by silencing gene tra-2 in the parental females via the pRNA interference. The data showed that tra-2 was transiently silenced in the female for three weeks period. The progenies sex ratio produced by these females during the silencing of tra-2 depart from 1:1 in favor of males. The excess of males was due to the transformation of a fraction of genotypical female XX embryos into XX males, the so-called pseudomales, Individual F1 males from the offspring of treated females crossed to females from the stock, revealed that majority of them showed regular mating behavior and were fertile. However, no offspring was produced in the crosses by a fraction of males that have produced sperms, showed regular mating behavior but did not transfer sperms to the females. The data allow the conclusion that the absence of tra-2 in the eggs had impaired the self-regulation of the embryonic gene tra resulting in the transformation of XX embryos into pseudomales and also that these pseudomales are sterile. This effect may be useful improve more sustainable technologies for fruit fly control such as SIT.


Introduction
Some fruit fly species of the Tephritidae family are considered one of the most devasting insect pests of agriculture The species of economic and quarantine importance (ca.78 species) damage the horticulture industry worldwide, reaching losses of billions of US dollars every year . The most important genera include Anastrepha (Schiner) indigenous to the Americas, Bactrocera (Macquart) in Asia and Oceania, Dacus (Fabricius) in Africa, Rhagoletis (Loew) in more temperate areas of Europe, North and South America, and the Zeugodacus (Hendel) in Asia (White & Elson-Harris, 1992;. The genus Anastrepha is endemic to the American tropics and subtropics, and many of its pest species have a wide distribution within this range (Norrbom & Foote, 1989). The South American fruit fly, Anastrepha fraterculus (Wiedemann) is a complex of cryptic species that has a wide geographic distribution almost the same as the genus and uses many host plants for larval development (Zucchi & Moraes, 2008;Norrbom et al., 2013). This species involves a complex of cryptic species, the so-called Anastrepha fraterculus complex of cryptic species that is being currently characterized (Selivon et al., 2004(Selivon et al., , 2005Hernández-Ortiz et al., 2012Hendrichs et al., 2015;Vaníčková et al., 2015;Prezotto et al., 2019). Due to the economic importance of the Anastrepha species programmes for their control, SIT for example, have been established in several countries along the continent . However, the effectiveness of such programs can in some cases be increased with the establishment of so-called sexing system to eliminate females allowing male-only releases. Therefore, new strategies to improve the massive production of males in the biofabrics are ever since being pursued, one of which is based on the knowledge of the genetic/molecular mechanism of sex determination of  (Pane et al., , 2007Salvemini et al., 2009), Bactrocera (Chen et al., 2008;Peng et al., 2015), Anastrepha (Sarno et al., 2010;Schetelig et al., 2012) as well as of MoYdsRNA in C. capitata and species of Bactrocera (Meccariello et al., 2019), results in the male-specific splicing of pre-mRNA of tra and of dsx in genotypic female embryos. This leads to transformation of a fraction of females into XX male individuals, the pseudomales. Moreover, functional role of the three genes, tra, tra-2 and dsx of Anastrepha, was also demonstrated by introduction of cDNA of the Anastrepha genes in intersexual Drosophila melanogaster and the development of sexually dimorphic structures were monitored. It was shown that genes tra (Ruiz & Sanchez, 2010), tra-2 (Sarno et al., 2011) and dsx (Alvarez et al., 2010) of Anastrepha promoted the transformation Drosophila XY male flies into females, i.e., the Anastrepha genes have changed the male splicing of the Drosophila pre-mRNAs into the female mode of splicing corroborating the role of the genes in Anastrepha sex determination. Similarly, it was demonstrated that the dsx gene of Ceratitis capitata induces masculinization of transgenic XX Drosophila flies (Saccone et al., 2008).
In the embryonic RNAi experiments with tephritid flies it was also detected that the transformation of females into XX pseudomales was usually not complete, and that fertility of pseudomales was also affected. In the XX pseudomales asymmetric and/or disformed testes and gynandromorphs were detected in Ceratitis capitata Salvemini et al., 2009), Bactrocera dorsalis , and in Anastrepha sp.1 aff. fraterculus (Sarno et al., 2010). Although the XX pseudomales of Anastrepha suspensa had motile sperms, they were shown to be sterile (Schetelig et al., 2012). On the contrary, the XX pseudomales of Ceratitis capitata (Salvemini et al. 2009, Meccariello et al 2019 and of Bactrocera dorsalis  were fully fertile. Most of sex determination experimental studies in tephritid fruit flies have been performed by injection of dsRNA into eggs which suppresses the function of the involved genes in the embryos. However, although sex determination also involves maternal components of genes tra and tra-2 in eggs Ruiz et al., 2007aRuiz et al., , 2007bSarno et al., 2010;Luo et al., 2017), no experimental demonstration of their function has been made. The present report describes the results of silencing the tra-2 expression in adult females demonstrating that the tra-2 mat is essential for the sex determination hierarchy in Anastrepha sp.1 aff. fraterculus.

dsRNA Preparation and Injection
For generating Aftra2dsRNA, tra-2 cDNA fragments were produced using Superscript ® III First-Strand Synthesis for RT-PCR kit from total RNA (5 µg per reaction), following manufacturer instructions using tra-2F primer. Afterwards, those fragments were amplified by PCR reaction using the following primers: tra-2F: ATGAGTCCACGTAC, tra-2R: CACGTCGCTTATCGTACGGA. The PCR was performed under the following conditions: 94 ºC for 1 min, followed by 35 cycles at 94 ºC for 20 sec, 55 ºC for 20 sec and 72 ºC for 30 sec, followed by a final extension step of 72 ºC for 10 minutes. PCR products were purified using GeneJET PCR Purification Kit (Thermo) and then sequenced using ABI Prism Big Dye Terminator Cycles Sequencing Ready Reaction kit, both following manufacturer instructions. Obtained sequences were manually inspected in BioEdit Sequencies Alignment Editor (v.7.0.9.0). Purified fragments were used in a second PCR using tra-2F and tra-2R primers containing T7 promoter sequences, T7tra-2F: ATGAGTCCACGTAC TAATACGACTCACTATAGGGAGA and T7tra-2R: CACGTCGCTTATCGTACGGATAATACGACTCACTATAGGGAGA. The PCR cycling was as following: 94 ºC for 1 min, followed by 35 cycles at 94 ºC for 20 sec, 65 ºC for 30 sec and 72 ºC for 30 sec, followed by a final extension step of 72 ºC for 10 minutes. Product samples were analyzed by electrophoresis in 1% agarose gels in 1x TAE buffer with 1x GelRedTM Nucleic Acid Gel Stain, 1x using TAE as running buffer for 30 minutes at constant 60V and 60mA, as recommend by Sambrook et al. (1989). Electrophoresis gels were analyzed under UV transilluminator, recorded (Viber-Laumat) and transferred to computer for analysis. After confirming the presence of the fragments, DNA was measured by NanoDropTM 1000 Sepectrophotometer (Thermo Scientific), and this product was used for dsRNA synthesis.
About 1µg of the product was used to produce Aftra2dsRNA by in vitro reaction with T7 polymerase for 16 hours using Megasript kit (Ambion). dsRNA products were purified by precipitating in lithium chloride (LiCl), following manufacturer instructions, and resuspended in injection buffer (5 mM KCl, 0.1 mM NaH 2 PO 4 , pH 6.8) to a final concentration of 2 µg/µl, measured with NanoDropTM 1000 Spectrophotometer (Thermo Scientific). Samples were used immediately or stored at -80 ºC.
For injection, solution of Af1tra2dsRNA at the concentrations of 1.8 µg/µL, 1.2 µg/µL and 0.6 µg/µL were prepared in buffer containing 5 mM KCL, 0.1 mM, NaH 2 -PO 4 , pH 6.8 (Sarno et al., 2010). Groups of 10 adult females under cold anesthesia were injected with dsRNA with a fine tip glass needle (Eppendorf) at the ventro-lateral pleura between abdominal segments 3 and 4. Three volumes of dsRNA solution (1 µL, 0.7 µL and 0.5 µL) from each of the three concentrations were tested as well as control injections of plain buffer. The injected females were transferred to cages provided with water and food ad libitum and inspected for the next 24 h. The female survival was higher (70 to 80%) at the dsRNA concentration of 1.2 µg/µL in a volume of 0.7 µL, similarly to the survival of females injected with plain buffer. These parameters for the dsRNA injection were employed in all the experiments. Number of flies, as well as experiment details and statistical methods of analysis will be explained in the Results for the sake of clarity.

Experimental Design
The study comprises three sets of experiments: Firstly, tests for the demonstration that the Af1tra2dsRNA induces the transient silencing of gene tra-2 in the adult females; secondly, analysis of sex ratio of progenies produced by the treated females during the period of tra-2 silencing; third, analysis of the morphology, of karyotypes and fitness parameters of the F1 males and morphology of F1 females from these progenies.

Maternal tra-2 Silencing
Efficiency of the Af1tra2dsRNA in promoting tra-2 gene silencing was tested by RT-PCR. For that, 18 females were injected with Af1tra2dsRNA and an equal number with plain buffer (BC). The injected females were separated in groups of 3 for RNA extraction in days 0, 3, 6, 9, 12 and 15 after injection (0 corresponding to the day of injection). RNA was extracted from total adult flies and employed to produce cDNA using SuperScript III ® (Thermo Scientific) following the fabricant instruction with tra-2 specific primers (F: 5' AGAGTTGGAATGAGTCCACGTAC 3'; R: 5' CACGTCGCTTATCGTACGGA 3). Primers were constructed using NCBI Primer-Blast tool, from tra-2 cDNA of Anastrepha obliqua (GenBank FN658607). For the RT-PCR reaction 10% of cDNA was employed and the cycles were as follows: 1 minute at 94 ºC, followed by 32 amplification cycles at 15 seconds at 94 ºC, 15 seconds at 57 ºC and 30 seconds at 72 ºC, and finally and final extension step of 10 minutes at 94 ºC. Then, the reactions were analyzed by running gel electrophoresis (agar 1.5% in TAE buffer) stained with GelRed ® . Gel runs were analyzed under UV transilluminator (Viber-Loumat), photographed and images were edited by Photoshop 6.0 (StatSoft, Inc ® ).

Analysis of Sex Ratio
Four replicate experiments (A, B, C, D) for testing the effect of dsRNA, and two controls, one with plain buffer (BC) and the other without any treatment (UC) were made. In each experiment, pupae of A. sp.1 aff. fraterculus from colony were transferred to cages until emergence of adults. Soon as they emerged males and females were separated in individual cages and provided with water and food ad libitum. For analysis of sex ratio, 50 pairs of mature virgin flies (15-17 days) were transferred to standard insect rearing cubic cages (20 cm edge), and guava fruits were provided for oviposition and larvae development. After a week, guavas were transferred to flasks, containing vermiculite. After 15-20 days, pupae were collected and maintained under standard condition until emergence of adults. Analysis of imagoes from each progeny revealed the sex ratio produced by females before of dsRNA and buffer (BC) treatments. At the end of this one-week pre-injection period, females of each replica were injected with Af1tra2dsRNA, others were injected with buffer (BC) and a non-injected untreated group of females (UC) was secured. To each group guavas were furnished for oviposition during a week to obtain progenies and this procedure was repeated for the next four weeks. The sex ratio of emerged adults produced at each experimental and control group during each of the five weeks the experiment lasted was registered. Additionally, groups of F1 males from the offspring of the experimental replicas were secured along the experiments for karyotype analysis of their as well as for their mating behavior, fertility, morphology of reproductive structures, production and transfer of sperms. Morphology of F1 females was also analyzed.

Karyotype Analysis
For this analysis, 40 F1 males derived from progenies of dsRNA injected females of each experimental replica and 40 from the control group (BC) were karyotyped as described by Selivon and Perondini (1967). Briefly, the jas.ccsenet.org Journal of Agricultural Science Vol. 13, No. 10; testicles of recent emerged males (1 to 2 days after emergence) were dissected, transferred to a 0.2% solution of sodium citrate for 10 min, immersed in a drop of 2% acetic orcein, and squashed between slide and coverslip. Analysis of the preparations was made under a BX60 Olympus microscope under regular illumination or phase contrast. Digital images were captured by a DC100 Leica camera coupled to the microscope and edited by Photoshop 6.0 (StatSoft, Inc. ® ).

Morphology and Reproductive Performance of Males
For these tests, 20 F1 males at the 1 st week and 20 at the 3 rd week from each experimental replica and from the control group (BC) were individually transferred to cages together with 2-3 virgin females from the stock. These cages were provided with water, food, and guavas as oviposition site for two weeks. The guavas from each replica were retrieved and maintained until pupae recovering. After emergence, adults were employed for in cross fertility tests. During this period, 5 males from replica A, 5 from replica B and 6 from replica C, died for unknown causes. Hence, the analysis of F1 male characteristics were based on 35 males from each experimental replica A and B, and 34 from replica C.
The males from each cross were observed daily for two weeks and their behavior related to courtship and mating were registered: everted anal pouch membranes (indicative of pheromone emission), wing fanning and copulas or copula attempts (Gomes-Cendra et al., 2011). After this period, the guavas were retrieved, maintained until adult emergence, and the presence of progenies derived from these fruits indicated fertility of the crosses.
After these two weeks, the F1 males were retrieved, anesthetized under cold and examined in a stereomicroscope for the external morphology of their terminalia. After, they were dissected for analysis of their internal reproductive organs. The testicles were removed and squashed between slide and coverslip in a drop of insect Ringer and examined for the presence of motile sperm. The external terminalia of the females involved in these crosses were also examined. After female dissection, their spermathecae were removed, squashed in drop of insect Ringer under a slide and coverslip, and examined for the presence of sperms. Hence, it was ascertained whether each one of the males was fertile or not and this information correlated with their morphological and other biological parameters analyzed. These analyses were made in a BX60 Olympus microscope under phase contrast. Digital images were taken by a DCC Leica camera coupled to the microscope and edited by Photoshop 6.0 software (Statsoft Inc. ® ).

Maternal tra-2 Silencing
In females injected with buffer (BC control) amplification of a fragment of about 300 bp was detected along three weeks after injection with no visible differences in the size of the tra-2 fragment compared to the fragment amplified from non-treated control females (UC). The fragments were sequenced and was confirmed that they correspond to tra-2 sequences (Sarno et al., 2010) (data not shown). In Af1tra2dsRNA injected females, amplification of a fragment of 300 bp was observed at the day of injection (day 0). However, on the 3rd day after injection a faint band was still obtained, followed by a period in which no amplification of tra-2 was detected, but on day 15 a faint fragment was detected again, indicating that expression of tra-2 was being resumed ( Figure  2). jas.ccsenet.

Af1tra2d
and the femal amplified

Sex R
The effect the four re produced b Figure 3). injected w males that about 1:1 although s differences females fr from the 1 of males in and 4 was the progen progressiv ng of gene tra-) in females inj ment of tra-2 w ment of 300 bp o day 12, but a for clarity. The RNAi were ana B, C, D) and tw eviously to the ary, the sex ra RNA has devi 50% before tre nd 5 ( Figure 3A ces between re distributions rol groups, BC e 1 and Figure he overall aver Figure 3B).  been resumed. This apparent inconsistence may be explained, considering that the eggs collected at day 21 had initiated their development 7-10 days before, thus during the period that tra-2 was still silenced in the parental female, as indicated in Figure 3B. Moreover, the fact that absorption of substances by the developing oocyte occurs only during the initial stages of their development (vitellogenesis stage), or in other words, before the stages the egg covers, vitelline and chorion membranes, are deposited in the oocyte (Telfer, 2009), may also be an explanation for the observed results.
In species of tephritid fruit flies, as in other insect species (Sánchez, 2008;Verhlust et al., 2010;Bopp et al., 2014), the increase in male frequency following injection of tradsRNA and/or tra2dsRNA in embryos cause the default expression of zygotic tra leading to the development of a fraction of genotypic XX embryos to development as males, the so called XX pseudomales (Salvemini et al., 2009;Sarno et al., 2010;Schetelig et al., 2012;. It shall made clear that in the present study the XX pseudomales were identified by chromosome analysis in a sample of F1 males derived from the sex ratio analysis. However, the karyotypes of another sample of 104 sister F1 males could not be determined because they were employed in the crosses to evaluate their reproductive status. Thus, one may assume that among these 104 F1 males a fraction must have had the XX chromosome pair while the majority were regular males bearing XY sex chromosomes. The analysis of these 104 F1 males showed they exhibited the typical mating behavior of the species herein analyzed (Selivon et al., 2005;Roriz et al., 2017). Hence, if pseudomales of A. sp.1 were present, they would show similar biological characteristics as the pseudomales of A. suspensa that although displaying regular mating behavior and producing motile spermatozoids, are sterile (Schetelig et al., 2012). Another aspect suggestive that the pseudomales herein produced were sterile is the fact that single sex progenies composed only by females were not found among the 72 fertile crosses of the present analysis. These only female offsprings would be produced by XX pseudomales if they were fertile as was observed in C. capitata (Savemini et al., 2009) and B. dorsalis . Hence, as pointed out by Schetelig et al. (2012), it seems that in C. capitata and B. dorsalis male fertility is not dependent of the Y chromosome while in A. suspensa the Y chromosome may carry factor(s) related to fertility. The results herein obtained for A. sp.1 aff. fraterculus, seems to be in line with these inferences.
Sterile crosses in which sperms were not transferred may be related to failure to complete the copula or to alteration in the female receptivity to mate, as shown in A. sp.1 from Argentina (Abrahams et al., 2011).
Since every male from the offspring of treated females, including the existent XX ones, was able copulate it was not expected morphological alterations in their terminalia, a fact that indeed was observed, as well as the presence of mobile sperms. Moreover, in the present study some of these males had altered testicles, usually hypertrophied ones and these characteristics were similar to those observed for males that develop from eggs of A. suspensa in which gene tra-2 had been silenced by injection of tra2dsRNA (Schetelig et al., 2012). Another morphological alteration in the progenies of treated females is a rare case of a female that had a deformed ovipositor. This morphological malformation seems to be similar to that described in B. dorsalis treated by parental dsxdsRNA in which a fraction of the F1 females showed shorter ovipositor (Chen et al., 2008).
Hence, in the present study it is assumed that in the absence of tra-2 mat in XX embryos the auto-regulatory expression of the zygotic tra were not activated and truncated/default Tra protein must have been produced leading XX embryos to follow male development. Clearly, regular XY embryos are not affected by the treatment given origin to regular male flies. It must be noted that the sex ratio deviation in favor of males was not caused by differential mortality of females as attested by the rate of adult emergence in progenies of the treated females (Table 2) but was due to the transformation of a fraction of genotypic XX embryos into pseudomales. This effect may be useful to generate improvements in the mass rearing of flies, since sexing of the insects is a significant bottleneck for the application of the SIT to insect pest. The production of pseudomales may be enhanced by association of dsRNA of the genes tra and tra-2 as reported for A. suspensa (Schetelig et al., 2012), and the dsRNAs for adult females could be delivered by other methods more suitable for the large-scale application in the biofabrics (Katoch et al., 2013;Zhang et al., 2013;Yu et al., 2013;Pomerantz & Hoy, 2015;Song et al., 2017). Improvements in modern and more sustainable technologies, as SIT, integrated with other control methods may lead to a more environmentally friendly strategies for insect pest control (Liedo et al., 2021).

Conclusion
The present report describes the results of the transient silencing of gene transformer-2 by parental dsRNA interference methodology in A. sp.1 aff. fraterculus. It is concluded that the tra-2 mat in the egg is a necessary factor for the auto-regulatory activity of embryonic gene transformer, the key gene for the sex determination hierarchy in Anastrepha. This effect may be useful to improve more sustainable technologies for fruit fly control such as SIT.