Karyotype variations in Italian populations of the peach-potato aphid Myzus persicae (Hemiptera: Aphididae)

Abstract In this study, we present cytogenetic data regarding 66 Myzus persicae strains collected in different regions of Italy. Together with the most common 2n = 12 karyotype, the results showed different chromosomal rearrangements: 2n = 12 with A1–3 reciprocal translocation, 2n = 13 with A1–3 reciprocal translocation and A3 fission, 2n = 13 with A3 fission, 2n = 13 with A4 fission, 2n = 14 with X and A3 fissions. A 2n = 12–13 chromosomal mosaicism has also been observed. Chromosomal aberrations (and in particular all strains showing A1–3 reciprocal translocation) are especially frequent in strains collected on tobacco plants, and we suggest that a clastogenic effect of nicotine, further benefited by the holocentric nature of aphid chromosomes, could be at the basis of the observed phenomenon.


Introduction
Classical and molecular cytogenetics provide an integrated approach for structural, functional and evolutionary analyses of chromosomes. This ranges from karyotype analyses to molecular mapping of chromosomes.
To date, studies concerning chromatin structure and organization have been mainly focused on eukaryotes having monocentric chromosomes, whereas species possessing holocentric/holokinetic chromosomes have been rather neglected. Chromosomes with diffused centromeric activity have been found in Protista, as well as in plant and animal species (Wrensch et al., 1994). The chromosomes of aphids, like those of other hemipteran insects, have diffuse centromeres so that kinetic activity is dispersed along the entire length of each chromatid at least in mitotic divisions, thus influencing chromosome behaviour (White, 1973). In organisms possessing this kind of chromatin organization, chromosome fusions and fissions can occur without any duplication or loss of centromeres. This has consequences for the survival of the de novo chromosomal changes through mitosis and meiosis, and hence for karyotype evolution. Autosomal fusions and fissions, particularly the latter, seemed to play a pivotal role in aphid karyotype evolution (Blackman, 1980), although this view is at present somewhat speculative due to a lack of knowledge concerning the mechanisms involved in rearrangements of the holocentric chromosomes (Spence & Blackman, 2000).
The standard female karyotype of this species is 2n = 12, but specimens with a chromosome complement of either 2n = 13 or 14 have also been reported (Blackman, 1980;Lauritzen, 1982). On the basis of relative chromosome lengths, Blackman (1971) concluded that the 2n = 13 karyotype raised from a break in one autosome of the pair A3, whereas a break in one chromosome of either the A2 and A3 pairs led to a 2n = 14 karyotype. Rare cases of strain possessing 2n = 11 and 3n = 18 have also been reported (Blackman, 1980;Yang & Zhang, 2000). Very recently, the analysis of mitotic metaphase chromosomes of a M. persicae laboratory strain revealed different chromosome numbers, ranging from 12 to 17, within each embryo (intraclonal genetic variation sensu Loxdale & Lushai (2003)). Chromosome length measurements revealed that the observed chromosomal mosaicism is due to recurrent fragmentations of chromosomes X, 1 and 3 (Monti et al., 2012).
The present study shows cytogenetic data regarding 66 M. persicae strains collected in different Italian regions showing several chromosomal rearrangements, the most
For chromosome spreads, adult females were dissected in Ringer saline solution and embryos were kept in a 1% hypotonic solution of sodium citrate for 30 min. The embryos

Karyotype variations in Myzus persicae
were then transferred to minitubes and centrifuged at 350 g for 3 min. Methanol-acetic acid 3:1 was added to the pellet, which was made to flow up and down for 1 min through a needle of a 1 ml hypodermic syringe to obtain disgregation of the material followed by a further centrifugation at 1000 × g for 3 min. This step was repeated with fresh fixative. Finally, the pellet was resuspended in new fixative, and 20 μl of cellular suspension was dropped onto clean slides and stained with 5% Giemsa solution in Soerensen buffer, pH 6.8 for 10 min. Silver staining of nucleolar organizing regions (NORs) was achieved following Howell & Black (1980). Slides were examined using a Nikon Eclipse 80i fluorescence microscope with UV filters, and photographs were taken using Nikon digital sight DS-U1. Morphometric analyses of mitotic plates were carried out on 30 metaphases using the software MicroMeasure, freely available at the Biology Department at Colorado State University website (http://rydberg.biology.colostate.edu/ MicroMeasure). Male induction for Salerno 03, Pescara 02, Cosenza 02 and Pisa 01 strains was evaluated by exposing parthenogenetic female aphids to short photoperiods (8 h light:16 h dark) according to Crema (1979).

Results
The analysis of mitotic cells of embryos, obtained from parthenogenetic females, confirmed that 2n = 12 is the standard chromosome number in M. persicae ( fig. 2), but 14 out of 66 strains analysed showed intraspecific karyotype variants due to both structural and numerical variations in chromosome complements (table 1, figs 3-6).
The most frequent chromosomal rearrangement found in Italian populations is related to the A1-3 reciprocal translocation, which was found either alone ( fig. 3) or together with an A3 fission (in one strain; fig. 6a, b). Other chromosome fissions involved A3 (found in two cases; fig. 4) and A4 (found in three cases; fig. 5), whereas a strain possessing 14 chromosomes as a consequence of both X and A3 fissions was also found ( fig. 6c,  d). Lastly, we identified a strain showing an intra-individual chromosome mosaicism due to the presence of mitotic plates with 12 (24% of the observed plates) and 13 (76%) chromosomes as a consequence of an A3 fission ( fig. 4b).
NOR staining (figs 3a, c, g, h and 6c) revealed the presence of heteromorphism in the size of rDNA genes in strains Salerno 3 ( fig. 4c) and Cosenza 2 ( fig. 6c) and evidenced that the fission of the X chromosomes observed in Cosenza 2 always occurred in the X chromosome bearing the smallest NOR-positive telomere and involved the X telomere opposite to the rDNA-bearing one ( fig. 6c).
Considering the geographical distribution, it is evident that almost all karyotype variations (11 out of 14) were present in central and southern Italian regions, whereas only three were found in northern locations. Furthermore, all but one of the strains collected on tobacco showed chromosomal rearrangements; and, in particular, all the strains possessing the A1-3 reciprocal translocation were found on this plant and were red in colour.
Male induction revealed that the M. persicae strains Salerno 03, Pescara 02 and Cosenza 02, all possessing different kinds of karyotype variations, are anholocyclic since it was not possible to induce the sexual generation differently from that obtained under the same experimental conditions with the M. persicae strain Pisa 1, which showed a normal karyotype.

Discussion
The typical aphid karyotype consists of pairs of rod-like chromosomes, whose number is typically stable within a genus, as shown in the large genus Aphis, where the typical chromosome number is eight with the exception of A. farinosa with 2n = 6 (Blackman, 1980;Hales et al., 1997). Nevertheless, exceptions have been published as revealed in the genus Amphorophora, where the chromosome number varies from 2n = 4 to 2n = 72 (Blackman, 1980).
Rearrangements most commonly involved autosomes, as shown in M. persicae, where, despite a standard chromosome number of 2n = 12, several strains possessing karyotypes consisting of 11-14 chromosomes have previously been reported (Blackman, 1980). On the contrary, Hales (1989) and Monti et al. (2012) demonstrated a complex pattern of associations and fissions occurring on both autosomes and X chromosomes in Schoutedenia lutea (van der Goot) (Hemiptera: Aphididae) and M. persicae, respectively, suggesting different scenarios for understanding aphid karyotype evolution.
The most common chromosomal variant described in M. persicae complement is a reciprocal translocation between the first and the third autosome pairs, leading to females with 2n = 12 karyotype showing a marked structural heterozygosity (Blackman, 1980). The empirical data, as presented in this paper, reveal for the first time that this chromosomal aberration also occurs in Italy since seven strains showed karyotype variations due to the A1-3 reciprocal translocation. In view of the absence of any primary constriction, which is typical of the holocentric chromosomes, together with the lack of specific banding patterns after conventional banding procedures, we combined procedures of standard chromosome staining (such as Giemsa and silver staining) with chromosome length evaluation. In particular, we used silver staining to confirm the exclusive localization of NORs regions on X chromosome telomeres in M. persicae and analyzed the involvement of sex chromosomes in the translocation event (Manicardi et al., 2002). Afterwards, in the absence of any other cytogenetic markers, the morphometric analysis was employed to identify autosomes A1 and A3 as the chromosomes engaged in the rearrangement.
According to the literature, a link exists between the A1-3 chromosomal reciprocal translocation and resistance to organophosphate and carbamate insecticides due to E4 gene amplification (Blackman et al., 1995), perhaps involving the removal of a repressor gene away from the structural genes in controls (Blackman et al., 1978). Preliminary data involving PCR and Southern blot analysis revealed that, in one of the Italian populations with this chromosomal aberration (Chieti 1), the FE4 gene (electrophoretically fast variant (allele) of the normal expressed carboxylesterase 4 (E4) enzyme) only was present (Rivi et al., 2009). This strain showed a moderate increase in esterase activity and was considered an S/R1 (susceptible/first resistance level) strain sensu Devonshire et al. (1992). The aforementioned data allows us to suggest that this is the first M. persicae strain possessing the A1-3 chromosomal reciprocal translocation linked to an FE4 and not directly related to a high level of esterase-based insecticide resistance. Experiments currently in progress are aimed to extend this experimental procedure to all Italian strains possessing A1-3 reciprocal translocations, in order to better clarify the relationships between this chromosomal rearrangement and the insecticide resistance in M. persicae populations.
Other fissions relatively frequent in the studied Italian M. persicae populations occurred at autosomes 3 and 4, whereas in one case only the fission involved the X chromosome. Different autosome fragmentations have been repeatedly described in M. persicae populations collected worldwide, whereas the X fragmentation has been observed only in a M. persicae laboratory strain characterised by an extensive chromosomal mosaicism (Monti et al., 2012). In this connection, it must be emphasized that in both such cases, the X fission occurs in X chromosomes possessing a low number of rDNA genes and in the telomeric region opposite to the NORsbearing one. The recurrent fission of the same chromosomes in the same region argues that the M. persicae genome possesses some fragile/labile sites that could be the basis for the observed changes in the chromosome number.
For many years, chromosome evolution has been generally explained by considering the random-breakage model (Becker & Lenhard, 2007). On the contrary, a number of comparative cytogenetic studies evidences a relationship between chromosomal rearrangements and specific chromosomal architecture and suggests a role of the repetitive DNAs in chromosome rearrangements. The nature of the repetitive DNA within chromosomal breakpoint regions varies significantly, from clusters of rRNA and tRNA genes to simple di-and tri-nucleotide expansions (Caceres et al., 1999;Carlton et al., 2002;Coghlan & Wolfe, 2002;Kellis et al., 2003;Renciuk et al., 2011). The data reported in this paper confirmed recent  observations regarding the recurrent fission of the same chromosomes in the same region (Monti et al., 2012), allowing us to further support the hypothesis concerning the presence of fragile/labile sites in the M. persicae holocentric chromosomes.
Chromosomal rearrangements in aphids have been hypothesized to affect some complex phenotypic traits, such as the host plant choice (Blackman, 1987;ffrench-Constant et al., 1988). For example, karyotypic variants observed in the corn leaf aphid Rhopalosiphum maidis (Fitch) have been associated with changes in the host choice. Similarly, an association of chromosome number with host plant has been described within the Sitobion genus, which shows 2n = 12 on ferns and 2n = 18 on grasses (Brown & Blackman, 1988;Hales et al., 1997).
A peculiar example of host adaptation concerns M. persicae strains feeding on tobacco. Morphometric analyses of specific taxonomic markers revealed that they are distinguishable from those living on other host plant so that the tobaccofeeding form was elevated to the status of a separate species by Blackman (1987). Further molecular evidences failed to confirm the genetic isolation of the population living on tobacco (Field et al., 1994;Clements et al., 2000), although other data, as well as behavioural/pheromonal evidence, suggest that the two forms undergone some significant degree of ecological-evolutionary divergence (Kephalogianni et al., 2002;Margaritopolous et al., 2003;Blackman et al., 2007).
Our data put in evidence that all but one of the strains collected on tobacco plants showed karyotype variations, whereas only four of the 56 population collected on other hosts (corresponding to about 7% of the total) displayed chromosomal rearrangements. A suggestive explanation for the observed relationships between chromosomal rearrangements and tobacco plants could rely in the clastogenic effect of nicotine.
Nicotine is a naturally occurring alkaloid found primarily in members of the solanaceous plant family, including Nicotiana tabacum. Several reports showed that nicotine, as a consequence of DNA replication fork stress (Richards, 2001;Freudenreich, 2005), produces genotoxic effects on Chinese hamster ovarian (CHO) cells (Trivedi et al., 1990(Trivedi et al., , 1993 and sister chromatid exchanges and chromosome aberrations in bone marrow cells of mice (Sen et al., 1991). Extensive chromosomal rearrangements have also been described in a mice population known as 'tobacco mice' since they live close to kiln for drying tobacco (Fraguedakis-Tsolis et al., 1997). In addition, DNA fragmentation by nicotine has been demonstrated both in peripheral lymphocytes (Sassen et al., 2005) and in human spermatozoa (Arabi, 2004). Nicotine, together with ultraviolet exposure, has also been considered an exogenous factor which can contribute to the generation of mutations which could be at the basis of chromosomal mosaicism (De, 2011), a very rare phenomenon we have observed in Salerno 02, one of the strains collected on tobacco plants.
Even if there are no literature data analyzing nicotine effects on organisms possessing holocentric chromosomes, the previously reported data allow us to propose at least that chromosome architecture, rather than random breakages, has a pivotal role in aphid chromosome evolution and rearrangements.
The high telomerase expression, previously reported in M. persicae (Monti et al., 2011), that stabilized chromosomes involved in fragmentations, coupled to reproduction by obligate apomictic parthenogenesis, could be at the basis of the stabilization of the observed chromosome instability on M. persicae strains collected on tobacco plants favouring the inheritance of the variant karyotypes.