Karyotypes, male meiosis and comparative FISH mapping of 18S ribosomal DNA and telomeric (TTAGG) n repeat in eight species of true bugs (Hemiptera, Heteroptera)

Abstract Eight species belonging to five true bug families were analyzed using DAPI/CMA3-staining and fluorescence in situ hybridization (FISH) with telomeric (TTAGG)n and 18S rDNA probes. Standard chromosomal complements are reported for the first time for Deraeocoris rutilus (Herrich-Schäffer, 1838) (2n=30+2m+XY) and Deraeocoris ruber(Linnaeus, 1758) (2n=30+2m+XY) from the family Miridae. Using FISH, the location of a 18S rDNA cluster was detected in these species and in five more species: Megaloceroea recticornis (Geoffroy, 1785) (2n=30+XY) from the Miridae; Oxycarenus lavaterae (Fabricius, 1787) (2n=14+2m+XY) from the Lygaeidae s.l.; Pyrrhocoris apterus (Linnaeus, 1758) (2n=22+X) from the Pyrrhocoridae; Eurydema oleracea (Linnaeus, 1758) (2n=12+XY) and Graphosoma lineatum (Linnaeus, 1758) (2n=12+XY) from the Pentatomidae. The species were found to differ with respect to location of a 18S rRNA gene cluster which resides on autosomes in Oxycarenus lavaterae and Pyrrhocoris apterus, whereas it locates on sex chromosomes in other five species. The 18S rDNA location provides the ﬁrst physical landmark of the genomes of the species studied. The insect consensus telomeric pentanucleotide (TTAGG)n was demonstrated to be absent in all the species studied in this respect, Deraeocoris rutilus, Megaloceroea recticornis, Cimex lectularius Linnaeus, 1758 (Cimicidae), Eurydema oleracea, and Graphosoma lineatum, supporting the hypothesis that this motif was lost in early evolution of the Heteroptera and secondarily replaced with another motif (yet unknown) or the alternative telomerase-independent mechanisms of telomere maintenance. Dot-blot hybridization analysis of the genomic DNA from Cimex lectularius, Nabis sp. and Oxycarenus lavaterae with (TTAGG)n and six other telomeric probes likewise provided a negative result.


Introduction
Fluorescence in situ hybridization (FISH), established in the 1980s, represents a powerful cytogenetic technique for a visualization of specific DNA sequences onto chromosomes, generating detailed chromosome mapping of eukaryote genomes (Pinkel et al. 1986). Despite the fact that the FISH mapping of insect chromosomes has been under way for a number of years (reviewed by Frydrychová et al. 2004;Vítková et al. 2005), the information of this sort for true bugs is still very scanty and available only for few species studied in respect to telomeric sequences (Okazaki et al. 1993;Sahara et al. 1999) and the location of ribosomal RNA genes (Cattani et al. 2004;Severi-Aguiar et al. 2005Papeschi and Bressa 2006;Morielle-Souza and Azeredo-Oliveira 2007;Bressa et al. 2008Bressa et al. , 2009Panzera et al. 2010;Grozeva et al. 2010; Bardella et al. 2010;Poggio et al. 2011).
To fill this gap and learn more about bug genomes, we applied FISH technique with telomeric (TTAGG) n and 18S rDNA probes to eight species belonging to 7 genera, 5 families and 2 infrforders: Deraeocoris ruber (Linnaeus, 1758), D. rutilus (Herrich-Schäffer, 1838), and Megaloceroea recticornis (Geoffroy, 1785) from the family Miridae, Cimex lectularius Linnaeus, 1758 from the family Cimicidae (all from the infraorder Cimicomorpha); Oxycarenus lavaterae (Fabricius, 1787) from the family Lygaeidae s.l., Pyrrhocoris apterus (Linnaeus, 1758) from the family Pyrrhocoridae, Eurydema oleracea (Linnaeus, 1758) and Graphosoma lineatum (Linnaeus, 1758) from the family Pentatomidae (all from the infraorder Pentatomomorpha). The 18S rDNA location provided the first physical landmark of the genomes of the species studied. The species D. ruber and D. rutilus were studied here for the first time likewise in terms of their standard chromosomal complements.
In five species, M. recticornis, D. rutilus, C. lectularius, E. oleracea, and G. lineatum, we used a (TTAGG) n telomeric probe to justify a hypothesis that this telomeric motif is absent in the true bugs (Frydrychová et al. 2004). The last hypothesis has been so far based only on studies of two species, Halyomorpha halys (Stål, 1855) (Okazaki et al. 1993: as Halyomorpha mista (Uhler, 1860) and Pyrrhocoris apterus (Sahara et al. 1999) that do not adequately represent the diversity of the Heteroptera.

Preparations
The gonads were dissected out and squashed in a drop of 45% acetic acid. The cover slip was removed using the dry ice. Slides were dehydrated in fresh fixative and air dried. The preparations were first analyzed with a phase contrast microscope at 400x. The best chromosome spreads were used for different staining techniques.

Fluorochrome banding
To reveal the base composition of C-heterochromatin, staining by GC-specific chromomycin A 3 (CMA 3 ) and AT-specific 4-6-diamidino-2-phenylindole (DAPI) was used according to Schweizer (1976) and Donlon and Magenis (1983) respectively, with some modifications. C-banding pretreatment was first carried out using 0.2 N HCl at room temperature for 30 min, followed by 7-8 min treatment in saturated Ba(OH) 2 at room temperature and then an incubation in 2xSSC at 60°C for 1 h. Furthermore, the preparation (without Giemsa) were stained first with CMA 3 (0.4 μg/ml) for 25 min and then with DAPI (0.4 μg/ml) for 5 min. After staining, the preparations were rinsed in the McIlvaine buffer, pH 7 and mounted in an antifade medium (700 μl of glycerol, 300 μl of 10 mM McIlvaine buffer, pH 7, and 10 mg of N-propyl gallate).

Fluorescence in situ hybridization (FISH) DNA isolation, PCR amplification, probe generation
Genomic DNA from a male of Pyrrhocoris apterus (Heteroptera, Pyrrhocoridae) was isolated using a Chelex-100 extracted method. FISH using a 18S rRNA gene probe was carried out on the chromosomes of D.  Table 2) from the genomic DNA of P. apterus, and labeled by PCR with biotin. Telomere probe (TTAGG) n was PCR amplified and labeled using primers TTAGG_F and TTAGG_R (Table 2) and Rhodamine-5-dUTP (GeneCraft, Germany).

Microscopy and imaging
Chromosome preparations were analyzed under a Leica DM 4000B microscope with a 100x objective. Fluorescence images were taken with a Leica DFC 350 FX camera using Leica Application Suite 2.8.1 software with an Image Overlay module. The preparations were stored partly at Institute of Biodiversity and Ecosystem Research, BAS in Sofia and partly at the Zoological Institute, RAS in St Petersburg.

Miridae
Deraeocoris ruber Linnaeus, 1758, 2n=30+2m+XY Fig. 1a-d The karyotype is described here for the first time. There are 16 bivalents, including a pair of very small and negatively heteropycnotic chromosomes taken as a pair of mchromosomes, and the univalent X and Y chromosomes which are largest and smallest chromosomes in the set, respectively . Diplotene and diakinesis stages were not detected, and bivalents displayed no chiasmata since meiosis is achiasmate. 18S rRNA genes were mapped on both sex chromosomes, the signals being more intensive on the Y. At spermatogonial metaphases, a number of very small intercalary signals could be in addition seen on the X (Fig. 1a).
Deraeocoris rutilus (Herrich-Schäffer, 1838), 2n=30+2m+XY Fig. 2a, b The karyotype is described here for the first time. It is much like that described above for D. ruber. Likewise, at PMI, there are 16 autosomal bivalents and the univalent X and Y chromosomes. One of the bivalents is very small, negatively heteropycnotic pair of m-chromosomes (Fig. 2a, b). The X is the largest and the Y is one of the smallest chromosomes in the set (excluding the m-chromosomes); autosomal bilvalents constitute a decreasing size raw. Diplotene and diakinesis stages were not observed, and meiosis was considered achiasmate of a collochore type.  FISH with an 18S rDNA probe produced two local signals placed near a telomeric region of the X chromosome and in addition a huge cluster of signals attached to the X; Y chromosome carried no signal (Fig. 2a, b). FISH with a TTAGG probe produced prominent hybridization signals (Fig. 2a, b, arrowed), which were occasionally located on the same chromosomes however most likely did not indicate the telomeres.

Megaloceroea recticornis
From the counts of 23 plates at prometaphase I (PMI), 15 autosomal bivalents and two univalent sex chromosomes, X and Y, were detected suggesting that this species displays 2n=30+XY ( Fig. 3a -c) in contrast to 2n=32+XY previously reported for this species in England (Leston 1957). The X is fairly large whereas the Y is one of the smallest chromosomes in the set, and the autosomal bilvalents constitute a decreasing size raw. There are no visible constrictions in the chromosomes, since they are holokinetic. During meiotic prophase, the diplotene and diakinesis stages escaped detection. At condensation stage, the bivalents showed no chiasmata however homologues were connected with each other by tenacious thread-like structures, at least, at one site, and the telomeric regions pushed off from each other (Fig. 3a, b). Taken together, the observations of meiosis suggest this species to display the achiasmate meiosis of a collochore type. Fluorescence in situ hybridization with a (TTAGG) n probe did not reveal positive signals on the telomeres although occasionally gave rise to variable interstitial hybridization signals, sometimes quite bright, on separate chromosomes (Fig. 3f ). Major ribosomal DNA cistrons were shown to locate on both X and Y chromosomes as detected by FISH with a 18S rDNA probe ( Fig. 3d-f ). At MII, X and Y chromatids associate forming an XY pseudo-bivalent ( Fig. 3f ) and segregate reductionally. MI plates are nonradial with X and Y chromosomes distributed among the bivalents (Fig. 3c), whereas MII plates are clearly radial, and XY pseudo-bivalent is located at the center of the ring formed by autosomes (Fig. 3f ).

Cimicidae
Cimex lectularius Linnaeus, 1758, 2n=26+X 1 X 2 Y not figured This study confirms that Cimex lectularius display 2n=26+X 1 X 2 Y as it was repeatedly reported previously (see Grozeva et al. 2010 and references therein). FISH with a TTAGG probe produced no signals on the chromosome spreads (not shown).

Lygaeidae s.l.
Oxycarenus lavaterae (Fabricius, 1787), 2n=14+2m+XY Fig. 4a, b In accordance with earlier published data (Grozeva 2004), the karyotype of this species includes 18 chromosomes as evidenced by a spermatogonial metaphase (Fig. 4a) and meiotic MI with 8 autosomal bivalents and univalent X and Y chromosomes (Fig. 4b). One of the bivalents is very small and negatively heteropycnotic and taken as a pair of m-chromosomes described earlier in six other species of the subfamily Oxycareninae (Grozeva 1995). The mchromosomes are likewise well recognized at the spermatogonial metaphase (Fig. 4a). The 18S rDNA signals could be easily seen on the second largest pair of autosomes (Fig. 4a, b).

Pyrrocoridae
Pyrrhocoris apterus (Linnaeus, 1758), 2n=22+X Fig. 5a-f In accordance with previously published data (Henking 1891, Ueshima 1979, Sahara et al. 1999, the species displays 2n=22+X in males as indicated by our observations of different stages of meiosis ( Fig. 5c-f ). FISH with an 18S rDNA probe produced clear interstitial signals on every homologue of a larger autosomal bivalent best demonstrated in Figures 5c and 5d. Figure 5b (meiotic prophase) shows that signals are present on   autosomes and absent on a sex chromosome body (arrowed). We call attention to difference in signal strength between the homologues (Fig. 5a, c, d), which is most likely caused by difference in 18S rRNA gene copy number.

Pentatomidae
Eurydema oleracea (Linnaeus, 1758), 2n=12+XY Fig. 6a-f This study confirms that E. oleracea has 2n=12+XY as previously reported by other researchers (Schachow 1932, Geitler 1939, Xavier and Da 1945. At different prophase stages and at MI (Fig. 6a-d), 6 autosomal bivalents and two univalent sex chromosomes, X and Y, were observed. The chromosomes are fairly large as compared with those in the above mentioned multichromosomal species. In E. oleracea, the X chromosome is medium-sized whereas the Y chromosome is the smallest in the set; autosomal bilvalents constitute a decreasing size raw. FISH with a (TTAGG) n probe did not reveal positive signals on chromosomal spreads. Clear 18S rDNA signals were evident on both sex chromosomes ( Fig. 6a-d). Results of fluorochrome staining were consistence with the FISH evidence since CMA 3positive/DAPI-negative regions were observed on the sex chromosomes confirming thus the presence here the ribosomal loci (Fig. 6e, f ). In meiosis, both MI and MII plates were radial with univalent sex chromosomes at MI (Fig. 6d) and an X and Y pseudo-bivalent at MII (not shown) being located at the centre of the ring formed by autosomes. A number of MI plates demonstrated the deviations from the radiality with some of the autosomal bivalents lying at the center of the ring or one of sex chromosomes lying outside the ring.

Graphosoma lineatum (Linnaeus, 1758), 2n=12+XY Fig. 7a-d
The karyotype of 2n=12+XY discovered here in G. lineatum is in accordance with that published previously for other populations of this species (Geitler 1939, Xavier andDa 1945). The karyotype closely parallels that in E. oleracea. The chromosomes are fairly large and noticeably larger as compared with the multichromosomal species. At MI, there are 6 bivalents and the univalent X and Y chromosomes, the X chromosome being medium-sized and the Y the smallest chromosome in the set; autosomal bivalents constitute a decreasing size raw ( Fig.  7a-d). FISH with a (TTAGG) n probe did not reveal positive signals on the chromosomal spreads. Ribosomal DNA cistrons were found to locate at the terminal position on the X as detected by FISH with an 18S rDNA probe (Fig. 7a, b) and DNA binding fluorochromes which revealed DAPI-dull/CMA 3 -bright bands on the X (Fig. 7c, d).

Dot-blot hybridization analysis
Dot-blot hybridization of the genomic DNA from Cimex lectularius, Nabis sp. and Oxycarenus lavaterae was performed using seven telomeric probes, ciliate (TTTTGGGG) n and (TTGGGG) n , nematode (TTAGGC) n , insect (TTAGG) n , shrimp (TAACC) n , vertebrate (TTAGGG) n and plant (TTTAGGG) n . All the experiments provided no hybridizing bands clearly suggesting some other molecular composition of telomeres in true bugs.

Standard chromosomal complements
We studied standard chromosomal complements of eight species from 6 genera and 5 families of the true bug infraorders Cimicomorpha ( On the other hand, Leston (1957) recorded M. recticornis (Miridae) in England as having 2n=32+XY; however this count was not corroborated by our observations of this species. The karyotype of M. recticornis in Bulgaria, as revealed in our work, is 2n=30+XY. We can not explain this incompatibility, especially as Leston provided neither photograph nor drawing of the chromosomal complement. It should be mentioned here that 2n=32+XY is the first whereas 2n=30+XY the second commonest karyotype in the Miridae (Kuznetsova et al. 2011). The chromosomal complements of D. rutilus and D. ruber were studied herein for the first time. These species were found to agree with one another in a karyotype of 2n=32+XY, with a pair of m-chromosomes among autosomes, and the karyotype formula is hence determined as 2n=30+2m+XY. A pair of chromosomes (the autosomes) known as m-chromosomes has been described in karyotypes of many bug species (Ueshima 1979). These chromosomes are typically extremely small, negatively heteropycnotic and behave differently as compared to autosomes and sex chromosomes during meiosis. However their origin and significance in genomes remain still obscure. The presence or absence of mchromosomes seems to represent a fairly stable character at higher taxonomic levels in the Heteroptera (Ueshima 1979). Until the present time, m-chromosomes have been discovered in as few as two Miridae species, Capsus ater (Linnaeus, 1758) and Dicyphus digitalidis Josifov, 1958(Nokkala and Nokkala 1986, Grozeva and Simov 2008, even though dozens Miridae species were studied in respect to karyotypes (see review: Kuznetsova et al. 2011). Thus, D. rutilus and D. ruber from our study increased the total number of mirid species with m-chromosomes to four. It is worthy of note that m-chromosomes were not described in the ten previously studied representatives of the genus Deraeocoris Kirschbaum, 1856 which species were shown to have 2n=32+XY as well (see Ueshima 1979). In some cases m-chromosomes might have been overlooked due to their too small size and negative heteropycnosis in meiosis (Kuznetsova et al. 2011). It remains to be added here that the species studied in this work display holokinetic chromosomes which lack primary constrictions (the centromeres) as in all other Heteroptera (Ueshima 1979).

Male meiosis
In all the seven species studied in this work, the first meiotic division is reductional for the autosomes and equational for the sex chromosomes, and vice versa -the second division is equational for the autosomes and reductional for the sex chromosomes. Such  (Schachow 1932, Geitler 1939, Xavier and Da 1945) Graphosoma lineatum 14 2n = 12+XY X chromosome 2n=12+XY (Geitler 1939, Xavier andDa 1945).
*Data from Grozeva et al. 2010 a behavior of sex chromosomes in male meiosis, or "post-reduction", as it is called, represents one of the unique cytogenetic characters of the Heteroptera being inherent in most bug species (Ueshima 1979). The species studied herein, all except Miridae species, showed the orthodox chiasmate meiosis in males with only one chiasma per bivalent, the meiotic pattern characteristic of holokinetic chromosomes (Halkka 1964, Nokkala et al. 2004). In common with several Miridae species studied so far in this respect ((Nokkala and Nokkala 1986;Grozeva et al. 2006Grozeva et al. , 2007Grozeva and Simov 2008a, b), the three mirid species from our work, Deraeocoris ruber, D. rutilus and Megaloceroea recticornis, were found to have achiasmate meiosis of a collochore type (best exemplified by M. recticornis). In meiosis of this type, diplotene and diakinesis stages are absent, and no chiasmata are formed between homologous which are however connected with each other, generally only at one site, by thread-like structures, the so-called collochores. The collochores have the function to hold homologous chromosomes together in the absence of chiasmata, and hence ensure their proper orientation and regular segregation at anaphase I. One of the distinctive properties of the true bug meiosis is a specific spatial arrangement of metaphase plates known as radial ones. Either at both metaphases, MI and MII, or at only one of those, the autosomes (either as bivalents at MI or as univalents at MII) form a ring with the sex chromosomes (either as univalents at MI or as a pseudo-bivalent at MII) lying in its center (Ueshima 1979). In two species studied here on this point, different patterns were observed. The mirid species Megaloceroea recticornis displayed MI plates nonradial with X and Y chromosomes distributed among the bivalents, and MII plates clearly radial with XY pseudo-bivalent located at the center of the ring formed by autosomes. Based on our observations of Eurydema oleracea, in this pentatomid species both MI and MII plates are radial. We emphasize however that MII plates in this species were more stable in this pattern compared to MI plates, which sometimes demonstrated the deviations from the radiality with some of the autosomal bivalents also lying at the center of the ring or one of sex chromosomes lying outside the ring. The differences between MI and MII in regard to their radial arrangement observed in E. oleracea are in agreement with the available data on species from other bug families, including the Pentatomidae (Rebagliati et al. 2003). In another pentatomid species, G. lineatum, the first metaphase was nonradial, however there was no MII plates to be analyzed.

Chromosomal location of 18S rDNA clusters
The nucleolus represents a subnuclear compartment of eukaryotic cells in which the synthesis of ribosomal RNA (rRNA) and formation of ribosomes take place (Busch and Smetana 1970). Nucleolar organizer regions (NORs) are usually detected in insects by silver nitrate (AgNO 3 ) and GC-specific fluorochrome (most commonly by CMA 3 ) staining. However silver treatment stains only active NORs (Hubbell 1985) being therefore inadequate to the study of NOR location onto chromosomes. In con-trast, fluorescence in situ hybridization (FISH) with rDNA probes directly detects the location of ribosomal RNA genes, regardless of their activity. In eukaryotes, 5S and 18S ribosomal genes (rDNA) are organized into two multigenic families, namely the major rDNA family formed by the 18S, 5.8S, and 28S genes and the minor one composed of 5S genes (Long and David 1980). Chromosomal mapping of genes is important for identification of chromosomes, which is especially difficult in groups of organisms with holokinetic chromosomes.
In this study we have characterized the chromosomal locations of 18S RNA genes in seven species from six genera of the families Miridae (Deraeocoris ruber, D. rutilus, Megaloceroea recticornis), Lygaeidae (Oxycarenus lavaterae), Pyrrhocoridae (Pyrrhocoris apterus), and Pentatomidae (Eurydema oleracea, Graphosoma lineatum). Data on all the species (as well as those on the whole families Miridae and Lygaeidae) were obtained for the first time bringing thus the total number of the species and genera studied to 30 and 18, respectively. The species were shown to exhibit different patterns of rDNA location. Some of the species showed their ribosomic cistrons located on sex chromosome, either on the X (D. rutilus) or on both X and Y (D. ruber, M. recticornis, E. oleracea, G. lineatum) whereas in other species (O. lavaterae, P. apterus) they were located on a pair of autosomes.
The location of 18S rDNA appeared different within the taxa, in which rDNA sequences were mapped in more than one species as in the mirid genus Deraeocoris, where D. rutilus displayed rDNA clusters concentrated on the X, but D. ruber on both X and Y chromosomes. These findings give no way of any inferences especially as a wide variation of chromosomal location for the major rDNA has been observed in different bug taxa, including variations between the co-generic species. For example, the genus Triatoma Laporte, 1832 from the subfamily Triatominae (Reduviidae), which is one of the most studied bug groups, clearly shows the interspecific variation (Severi-Aguiar et al. 2005, Morielle-Souza and Azeredo-Oliveira 2007, Bardella et al. 2010, Poggio et al. 2011) while sometimes even intraspecific variation (Panzera et al. 2010) for the major rDNA harboring either on the sex chromosomes (X and/or Y), or the autosomes or on both. This variability is suggested to be due to the chromosomal exchanges between the autosomes and sex chromosomes during the speciation of the Triatominae (Panzera et al. 2010).

The telomere repeat sequence
The pentanucleotide sequence (TTAGG) n is known as the commonest and most likely an ancestral DNA motif of insect telomeres (Sahara et al. 1999, Frydrychová et al. 2004. However this motif was lost during the evolution of several groups being secondarily replaced with another motif (yet unknown) or the alternative telomerase-independent mechanisms of telomere maintenance (Frydrychová et al. 2004, Lukhtanov and. The true bugs are considered as one of the insect higher taxa in which (TTAGG) n is absent, however the data available concerned so far the two species only, Halyomorpha halys (Stål, 1855) (Pentatomidae) studied using Southern hybridization (Okazaki et al. 1993: as Halyomorpha mista (Uhler, 1860) and Pyrrhocoris apterus (Pyrrhocoridae) subjected both to Southern hybridization and FISH (Sahara et al. 1999). Comparative analysis of the occurrence of (TTAGG) n in various groups of insects has showed that this motif is evolutionarily stable, and, having once appeared during evolution, marks taxa and phylogenetic branches of high rank. It is known however that in some groups, such as the orders Coleoptera and Neuroptera, both TTAGG-positive and TTAGG-negative species are encountered (Frydrychová et al. 2004 and references therein). By using FISH we studied the occurrence of (TTAGG) n telomere repeat in five species: Deraeocoris ruber and Megaloceroea recticornis (Miridae), Cimex lectularius (Cimicidae), Eurydema oleracea and Graphosoma lineatum (Pentatomidae). All these species were shown to lack the insect consensus sequence. Although in both mirid species a number of prominent hybridization signals could be seen on separate chromosomes, these signals most likely did not indicate the telomeres. The presence of signals suggests a sequence related to TTAGG but it seems to have no target specificity in the bug chromosomes.
The absence of (TTAGG) n telomeric repeat in the phylogenetically distant groups within the Heteroptera strengthens thus the view (Frydrychová et al. 2004) that it was lost in early evolution of this group of insects.
Dot-blot hybridization of the genomic DNA from bug species with seven types of telomeric probes, ciliate (TTTTGGGG) n and (TTGGGG) n , nematode (TTAGGC) n , insect (TTAGG) n , shrimp (TAACC) n , vertebrate (TTAGGG)n and plant (TTTAGGG) n , yielded negative results and did not provide hence any answer of the question which is the telomere repeat sequence in bug chromosomes.

Aknowledgments
This study was supported financially by the Russian Foundation for Basic Research (grant 11-04-00734) and programs of the Presidium of the Russian Academy of Sciences "Gene Pools and Genetic Diversity" and "Origin of the Biosphere and Evolution of Geo-biological Systems" (for VK and BA) and by National Scientific Fund of Bulgarian Ministry of Education, Youth and Science (DO-02-259/08) (for SG). We express our thanks to Nikolay Simov (NMNH, Sofia) who provided us the insects for present study.