Cytogenetic analysis of three species of Pseudacteon (Diptera, Phoridae) parasitoids of the fire ants using standard and molecular techniques

Pseudacteon flies, parasitoids of worker ants, are being intensively studied as potentially effective agents in the biological control of the invasive pest fire ant genus Solenopsis (Hymenoptera: Formicidae). This is the first attempt to describe the karyotype of P. curvatus Borgmeier, P. nocens Borgmeier and P. tricuspis Borgmeier. The three species possess 2n = 6; chromosomes I and II were metacentric in the three species, but chromosome pair III was subtelocentric in P. curvatus and P. tricuspis, and telocentric in P. nocens. All three species possess a C positive band in chromosome II, lack C positive heterochromatin on chromosome I, and are mostly differentiated with respect to chromosome III. P. curvatus and P. tricuspis possess a C positive band, but at different locations, whereas this band is absent in P. nocens. Heterochromatic bands are neither AT nor GC rich as revealed by fluorescent banding. In situ hybridization with an 18S rDNA probe revealed a signal on chromosome II in a similar location to the C positive band in the three species. The apparent lack of morphologically distinct sex chromosomes is consistent with proposals of environmental sex determination in the genus. Small differences detected in chromosome length and morphology suggests that chromosomes have been highly conserved during the evolutionary radiation of Pseudacteon. Possible mechanisms of karyotype evolution in the three species are suggested.


Introduction
The Phoridae, one of the most diverse families within the order Diptera, includes more than 26,000 min species, varying in life style from scavengers and predators to parasitoids (Disney, 1994;Gilbert and Jervis, 1998). The phorid genus Pseudacteon Coquillet are parasitoids, most of the species parasitizing Solenopsis Westwood fire-ants (Hymenoptera: Formicidae) (Disney, 1994). Pseudacteon flies are solitary parasitoids (Disney, 1994). The females inject a single egg into each worker ant, the third larval instar of which migrating into and later pupating within the host's cephalic capsule (Porter et al., 1995;Porter, 1998).
Despite numerous studies on the life-history traits of the genus Pseudacteon, including apparent environmental sex-determination in some species, cytogenetic studies are lacking. Dipteran species generally possess low diploid chromosome numbers, these ranging from 2n = 4 to 20, with modal numbers at 6 and 8 (due to several families of "lower" Diptera i.e. SO Nematocera) and 12 (with many species of the SO Brachycera and belonging to calyptrate and acalyptrate families) (White, 1973). Almost all species of Sarcophagidae and Calliphoridae (SO Brachycera, superfamily Oestroidea) have 2n = 12 with a pair of sex chromosomes XY (Parise-Maltempi and Avancini, 2000; long acrocentric chromosomes (including the sex chromosome pair XY) and a small dot-like pair. Reduction in chromosome number has occurred many times independently by fusion of these elements (White, 1973). Many derived sex chromosome systems have been described and the evolutionary mechanisms of sex chromosome differentiation has been analyzed in Drosophila species (Steinemann and Steinemann, 2005). In Muscidae (SO Brachycera, superfamily Muscoidea) most species also possess 2n = 12 with XY sex chromosomes in the male. However, muscid species have been described with 2n = 10 chromosomes and lacking differentiated sex chromosomes (Parise-Maltempi and Avancini, 2001). In Phoridae (SO Brachycera, superfamily Phoroidea), Megaselia scalaris Loew and M. spiracularis Schmitz, the only 2 species that have been studied cytogenetically, possess the diploid number 2n = 6, with 2 metacentric and 1 telocentric chromosome pairs. They do not possess heteromorphic sex chromosomes and sex is determined by a single Maleness factor that can be located in any of the three linkage groups (Mainx, 1964;Traut and Willhoeft, 1990).
In order to evaluate whether the well-marked differences among P. curvatus Borgmeier, P. nocens Borgmeier and P. tricuspis Borgmeier in size, morphology, behavior and phylogenetic relations are also reflected in their karyotypes, our aim was to describe the karyotype of these species by using standard and molecular cytogenetic techniques. Descriptive values of the karyotype, heterochromatin content and distribution, and the location of nucleolus organizer regions (NORs) were analyzed. A comparative analysis of cytogenetic results also provided information on the possible karyotype evolution within the genus. Furthermore, we analyzed the presence of differentiated sex chromosomes, since it has been suggested, in the case of other species in the genus Pseudacteon, that sex determination is related to host size, and thus could be environmentally determined (Morrison et al., 1999).

Slide preparation
Different oviposition assays were performed in the lab following the same methodology used by Chirino et al. (2009). The "attacked" ants were placed in a rearing room at 28 ± 1°C with a 12:12 (light: darkness) photoperiod and 80 ± 10% RH.
In Diptera, cytogenetic analyses are mainly carried out with third instar larvae or early pupae Cevallos and Nation, 2004). Since development in Pseudacteon takes place within its host (Porter et al., 1995), it is difficult to obtain larvae at the appropriate developmental stages for cytogenetic studies. Thus we dissected pupae of 3-5 days, which lacked differentiated tissues, but showed high mitotic indices.
Since the females of P. tricuspis and P. nocens emerge from larger ant-heads than males, parasitized worker heads were measured and developed pupae were classified as either potential females or males. In contrast, both sexes of P. curvatus develop in similar sized hosts, so they could not be classified by sex (Morrison et al., 1997;Folgarait et al., 2006).

Chromosome morphology
Suitable cells at the mitotic prometaphase were selected and the mean descriptive values of the karyotype were calculated using information obtained from at least 3 cells per slide for each individual of each species analyzed (20.78 ± 9.76 cells). Each slide represented different individual pupae. Mitotic prometaphases were obtained from P. tricuspis pupae (17 males and 24 females); from P. nocens pupae (27 males and 21 females), and 35 specimens of P. curvatus. Slides were stained with the fluorescent dye DAPI (Rebagliati et al., 2003) for morphological studies. The nomenclature of Levan et al. (1964) was used to describe chromosome morphology.

Banding techniques
Heterochromatin content and distribution was analyzed by means of the C-banding technique and sequential DAPI and CMA 3 banding. C-bands were performed according to Sumner (1972) with slight modifications. Slides were hydrolyzed in HCl 0.2N at room temperature for 15 min, briefly washed in distilled water and then incubated at room temperature for 35 min in a 5% Ba(OH) 2 solution. Slides were thoroughly washed in tap water for 1 min and incubated for 1 h in 2XSSC at 60°C. After washing in distilled water, the slides were stained with DAPI (Rebagliati et al., 2003), since band resolution improves when pretreated slides are stained with DAPI instead of Giemsa. Fluorescent staining with GC specific chromomycin A 3 (CMA 3 ) and AT specific 4'6-diamidino-2-phenylindole (DAPI) was carried out according to Rebagliati et al. (2003).
For location of NORs, Ag-NOR staining was applied according to Howell and Black (1980). Fluorescent in situ hybridization (FISH) with 18S rDNA was also undertaken.

Fluorescence in situ hybridization
Unlabelled 18S rDNA probes were generated by PCR, using primers described by Fuková et al. (2005) and sequenced using an Automated Sequencer ABI 3100 (DNA Sequencer, PE, Applied Biosystems), thereby confirming their being a section of 18S rDNA of Ceratitis capitata (Weidemann) (Diptera: Tephritidae). PCR was done in a Mastercycler ® Gradient Eppendorf thermal cycler (Eppendorf AG, Hamburg, Germany). Reactions were carried out with template genomic DNA extracted from C. capitata by standard procedure according to Baruffi et al. (1995). The PCR product showed a single band of about 900 bp on a 1% agarose gel. The band was recovered from the gel and purified by using Wizard SV gel and PCRclean up (Promega). The rDNA probe was labeled by nick translation with biotin 14-dUTP (BioNick Labeling System, Invitrogen Life Technologies Inc., Buenos Aires , Argentina). FISH with a biotinylated probe was carried out as described by Sahara et al. (1999) with several modifications as described by Fuková et al. (2005). Hybridization signals were detected with streptavidin-Cy3 conjugate (Sigma, Saint Louis, USA). The preparations were counterstained with 0.5 mg/mL DAPI in PBS, 1% Triton X-100, and mounted in anti-fade (Vectashield Mounting Medium, Vector Laboratories, Inc., Burlingame, USA).

Photographs
Preparations were observed in a Leica DMLB epifluorescence microscope. Black-and-white images of chromosomes were recorded with a CCD camera (Leica DFC350 FX, Leica IM50 Version 4.0, Leica Microsystems Imaging Solutions Ltd. Cambridge, UK) separately for each fluorescent dye. Images were pseudocolored (light blue for DAPI and red for Cy3), and, when necessary, superimposed with the aid of AdobePhotoshop version 6.0.

Statistical analysis
The following chromosome measurements were performed with MicroMeasure for Windows, version 3.3: centromeric index CI (length of short arm as a percentage of the whole chromosome), arm ratio r (relationship between the long and the short arms of each chromosome), relative length % (length of a chromosome as a percentage of the total chromosome length), and total chromosome length -TCL (the sum of the lengths of all the chromosomes of the complement). Comparisons of data within and between species were undertaken by analyses of variance (ANOVA), adjusted by the Bonferroni method. Statistical comparisons were done using the STATISTIX Program (Analytic Software 1998, Tallahassee, FL).

Results
Analysis of mitotic chromosomes of the pupae of Pseudacteon tricuspis, P. curvatus and P. nocens revealed a 2n = 6 diploid number (Figure 1 a-b). Chromosome pairs I and II were metacentric in the three species, but chromosome pair III was subtelocentric in P. tricuspis and P. curvatus, and telocentric in P. nocens (Table 1). Neither P. tricuspis nor P. nocens possessed a distinguishable heteromorphic pair in potential males or females (t = 0.43, df = 39, p = 0.67 for P. tricuspis and t = 0.07, df = 46, p = 0.95 for P. nocens, 2-tailed t-Test). The three species did differ in total chromosome length (TCL) (F 2,99 = 2.38, p = 0.0453) which was higher in P. curvatus, lower in P. nocens and intermediate in P. tricuspis (Table 2). Chromosome I in P. nocens was relatively larger than in P. tricuspis 742 Karyotype of Pseudacteon species and P. curvatus (F 2,123 = 8.68, p = 0.0004). Chromosome II in P. nocens was also larger than in P. curvatus (F 2,123 = 5.86, p = 0.0039), whereas in P. tricuspis it was of intermediate length. Finally, chromosome III was smaller in P. nocens than in P. tricuspis and P. curvatus (F 2,123 = 29.14, p < 0.0001) ( Table 2). Different banding techniques were applied in order to reveal heterochromatin and analyze its location and composition. DAPI and CMA 3 sequential fluorescent banding did not reveal any regions either in AT rich (DAPI bright) or GC rich (CMA 3 bright) for any chromosome pair in any of the three species (Figure 1 c-f). After C-banding treatment, slides were stained with DAPI, allowing a better analysis of the C-heterochromatin pattern ( Figure 2). C positive heterochromatin was scarce in all species, with a small C positive (C+) band in the long arm (q) of chromosome II and a completely C negative (C-) chromosome I (Figure 2a-f). "Somatic pairing" was evident from prophase up to meta-phase (Figure 2a, d-i), but was lost during anaphase ( Figure  2b-c). The three species varied as to the C-banding pattern in chromosome III. Thus, in P. curvatus a C+ band present in the short arm (p) extended over 81% of its length (Figure 2a-c). In P. tricuspis, a C+ band observed in the long (q) arm near the centromere corresponded to approximately 4% of its length (Figure 2d-e). Finally, the telocentric pair in P. nocens was completely C- (Figure 2f).
Silver impregnation revealed only one nucleolus in interphase nuclei and two signals in early prophase cells in the three species. However, no positive results were obtained in metaphase chromosomes (Figure 3). In situ hybridization with the 18S rDNA probe on metaphase cells showed a single hybridization signal in chromosome II in the three species at a location similar to that of the C+ band (Figure 2g-i; Figure 4). This suggests that in the three Pseudacteon species a single cluster of rRNA genes is located in the long arm (q) of chromosome II near the centromere.

Discussion
The Pseudacteon species analyzed presented 2n = 6, with chromosomes I and II being metacentric and chromosome III subtelocentric in P. curvatus and P. tricuspis but telocentric in P. nocens. Total chromosome length (TCL) varied among Pseudacteon species, with P. curvatus ³ P. tricuspis ³ P. nocens, probably representing variations in DNA content that do not affect chromosome morphology. A similar situation has been reported in other dipteran fami- Chirino et al. 743  The comparisons of chromosomal lengths and the relative lengths (%) were made by means of one-way ANOVA. Different letters indicate significant differences (p < 0.05) adjusted by Bonferroni. The comparisons of the total chromosomal lengths (TCL) and the relative lengths (%) were analyzed by means of one-way ANOVA. Different letters indicate significant differences (p < 0.05) adjusted by Bonferroni.
lies (Rafael and Tadei, 1998;Parise-Maltempi and Avancini, 2001;Selivon et al., 2005). Heterochromatin content was very scarce in these Pseudacteon species. Chromosome I was completely C-, with a single C+ band situated at the location of the 18S rDNA hybridization signal on chromosome II. Major differences were observed in chromosome III, with a C+ band in the p-arm of the subtelocentric pair in P. curvatus and the q-arm in P. tricuspis, although this was absent in the telocentric pair of P. nocens (Figure 4). The divergence between the former two species could be explained by pericentric inversion in chromosome III during their evolutionary history ( Figure 5). Its absence in P. nocens suggests that it could have diverged before the differentiation of P. tricuspis and P. curvatus. Furthermore, the smaller size of chromosome III in P. nocens, as compared to the other two species, could be explained by the absence of a heterochromatin block. Although it would be necessary to analyze other Pseudacteon species, we propose here a possible hypothesis on karyotype evolution within the genus. We propose that after the divergence of P. nocens, and through the acquisition of heterochromatin, there was an increase in DNA across the genome of the ancestor of the other two species, but principally in chromosome III. In spite of the fragmented and unresolved phylogenies of some Pseudacteon species (Kronforst et al., 2007;Calcaterra et al., 2008), several conclusions support our hypothesis. The recent common ancestor of P. nocens is different from the most recent common ancestor of P. curvatus and P. tricuspis (Calcaterra et al., 2008), and phylogenies based on the mitochondrial genes Cytochrome Oxidase I and II and the nuclear gene Wingless (Kronforst et al., 2007) suggest that P. tricuspis could have derived from the ancestor from which P. curvatus diverged.
In reference to the studied Pseudacteon species, Ag-NOR banding revealed the presence of one nucleolus but not on metaphase chromosomes. Similar observations have been reported in other dipterans in which results from silver staining were unsatisfactory, and only centromeric cores could be differentiated (Motara et al., 1985;Wallace and Newton, 1987;Bedo and Webb, 1989;Marchi and Pili, 1994).
Cytogenetic characterization of Pseudacteon species failed to reveal morphologically differentiated sex chromosomes. The only other two species in the Phoridae in which cytogenetical data has come under analysis were Megaselia scalaris and M. spiracularis. Accordingly, both possess the same diploid number 2n = 6 and a similar chromosome morphology, with two metacentric and one telocentric chromosomes (Mainx, 1964). In this genus sex determination depends on the presence of a single Maleness factor M that triggers off male development (Mainx, 1964;Willhoeft and Traut, 1990;Traut and Wollert, 1998;Traut et al., 1999). Through molecular cytogenetic techniques and molecular markers it has been established that the chromo-744 Karyotype of Pseudacteon species  somes X and Y in Megaselia are in a very early state of molecular differentiation (Traut, 1994).
In most of the studied Pseudacteon species, the sex of the offspring seems to be facultatively determined by the size of the host, with females emerging from the larger ones (Morrison et al., 1999;Folgarait et al., 2005aFolgarait et al., , 2006. From an evolutionary point of view, this makes sense since 1) fire-ant workers are highly variable in size (Tschinkel et al., 2003), 2) female flies have a higher fitness if they emerge from larger hosts, exhibiting a higher efficiency in development and larger sizes (Chirino, Gilbert, Folgarait, unpublished), and 3) females emerge from a narrower range of ant-sizes than males (Folgarait et al., 2005a(Folgarait et al., , 2006Chirino, Gilbert, Folgarait, unpublished). However, in two small species, P. cultellatus Borgmeier (Folgarait et al., 2002) and P. curvatus (Chirino et al., 2009), females and males develop in similar-sized hosts, and emerged adults are not sexually dimorphic in size, so that the above mentioned pattern is difficult to detect. The number of different species of Pseudacteon that develop on S. invicta exceeds the diversity observed in other ants (Disney, 1994). This diversity could be influenced by Solenopsis size polymorphism since there is a correlation between the sizes of both parasitoid and host (Morrison et al., 1997). Different Pseudacteon species oviposit at different times during the day, at different locations (close to mounds or the foraging trails of ants) and during different seasons. P. nocens in particular is crepuscular and abundant in shady conditions, but P. curvatus and P. tricuspis are more abundant at midday and in sunny conditions (Folgarait et al., 2007a;2007b). Furthermore, P. curvatus mainly attacks small workers (Chirino et al., 2009), P. tricuspis larger ants (Morrison et al., 1997), and P. nocens, medium-sized ones. Finally, P. tricuspis females exhibit competitive, aggressive and territorial behaviors, which are not observed in either P. curvatus or P. nocens. In this work we have shown that the well-known differences among P. curvatus, P. nocens and P. tricuspis as regards size, morphology, behavior and phylogenetic relations are also reflected in their karyotypes.
The exact mechanism of sex determination in Pseudacteon flies is perplexing (Porter, 1998). Environmental sex determination (ESD) has been suggested as a possible mechanism for the observed pattern (Morrison et al., 1999). Typically, taxa having non-differentiated sex chromosomes usually show labile sex-determination, and sex can be determined by environmental stimuli (Solari, 1994). Our demonstration of non-differentiated sex chromosomes in Pseudacteon phorids is thus consistent with the earlier suggestion of environmental sex determination in these parasitoid flies, and should encourage further investigation of how sex is determined in this genus.

Acknowledgments
We thank Lic. Gloria Albioni Montenegro for her help in several aspects of the experiments. This research was supported by grants from the Programa de Investigación en Interacciones Biológicas (PPUNQ-0340/03), Universidad Nacional de Quilmes; the R.J. Kleberg & H.C. Kleberg Foundation; Universidad de Buenos Aires (X164) and PICT 2007-00635 of ANPCyT. M.G.Ch., P.J.F. and Chirino et al. 745