Comparative cytogenetics of Auchenorrhyncha (Hemiptera, Homoptera): a review.

A comprehensive review of cytogenetic features is provided for the large hemipteran suborder Auchenorrhyncha, which currently contains approximately 42,000 valid species. This review is based on the analysis of 819 species, 483 genera, and 31 families representing all presently recognized Auchenorrhyncha superfamilies, e.i. Cicadoidea (cicadas), Cercopoidea (spittle bugs), Membracoidea (leafhoppers and treehoppers), Myerslopioidea (ground-dwelling leafhoppers), and Fulgoroidea (planthoppers). History and present status of chromosome studies are described, as well as the structure of chromosomes, chromosome counts, trends and mechanisms of evolution of karyotypes and sex determining systems, their variation at different taxonomic levels and most characteristic (modal) states, occurrence of parthenogenesis, polyploidy, B-chromosomes and chromosome rearrangements, and methods used for cytogenetic analysis of Auchenorrhyncha.

Introduction

The hemipteran (homopteran) suborder is divided into two major lineages: the infraorder with superfamilies (cicadas), (spittle bugs), (leafhoppers and treehoppers), and (ground-dwelling leafhoppers), and the infraorder with the single superfamily (planthoppers) (Szwedo et al. 2004, Aguin-Pombo and Bourgoin 2012). More than 42,000 valid species of have been reported worldwide (Deitz 2008), which, depending on the classification followed, can be grouped roughly into 30 to 40 families.

Olli Halkka (Halkka 1959), one of the earliest and most well-known researchers of chromosomes in , concluded that they “are a group well suited for comparative karyological work. Technically, this group presents no special difficulties. The numbers of the chromosomes are relatively low and the chromosomes themselves are fairly large”. The first cytogenetic studies on provided data for the cicada species Linnaeus () (Wilcox 1895) and the spittlebug species Say () (Stevens 1906). Shortly afterwards Boring (1907) initiated research on the comparative karyology of with a study of 22 species belonging to five families. Documented lists of chromosome numbers have been published by several authors. Those of Halkka (1959) and Kirillova (1986, 1987) cover the complete suborder . The first author discussed different aspects of auchenorrhynchan cytogenetics, while the second reported only chromosome numbers and sex determining systems. Later, lists for particular fulgoromorphan families were published by Kuznetsova et al. (1998) for , , , , , , and ; by Maryańska-Nadachowska et al. (2006) for , , and ; by Kuznetsova et al. (2009a) for and ; and then, Tian et al. (2004) added data for 19 species of the families , , , , , , and , while Kuznetsova et al. (2010) for 14 species of the issid tribe . In contrast to , the data on cicadomorphan families have never been tabulated after the comprehensive reviews by Halkka (1959) and Kirillova (1986, 1987). Quite recently, chromosome numbers were reported for 91 species of (Wei 2010, Juan 2011) and 25 species of (Tian and Yuan 1997) from China. Several additional species were also karyotyped within the families , , , , and (Marin-Morales et al. 2002, Perepelov et al. 2002, Kuznetsova et al. 2003, 2013, 2015a, Aguin-Pombo et al. 2006, 2007, Maryańska-Nadachowska et al. 2008, 2012, Castanhole et al. 2010, de Bigliardo et al. 2011, Golub et al. 2014).

At the present time, approximately 819 auchenorrhynchan species (nearly 2% of the total number of species described) are known from a cytogenetic viewpoint (V. Kuznetsova, unpublished checklist). These species represent 483 genera and 31 families from all the superfamilies of . Of these taxa, 511 species, PageBreak335 genera and 11 families belong to , while 308 species, 148 genera, and 20 families belong to (Figs 1–8). The available data were chiefly obtained using conventional cytogenetic techniques and concerned, almost entirely, chromosome numbers, sex determining systems, and, in outline, the behaviour of chromosomes during meiosis. A few recent studies have used modern cytogenetic techniques to identify the individual chromosomes in karyotypes and specific regions in chromosomes of auchenorrhynchan species (Kuznetsova et al. 2003, 2009b, 2010, 2015a, Maryańska-Nadachowska et al. 2008, 2012, 2013, Golub et al. 2014). The application of new techniques, primarily , opened a promising area of research, which yields more detailed karyotype information (Maryańska-Nadachowska et al. 2013, Golub et al. 2014, Kuznetsova et al. 2015a).

Figure 1.

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 308 species and 148 genera of the families , , , , , , , , , , , , , , , , , , , and .

Figure 8.

Histograms showing the distribution of female diploid chromosome numbers in (a, b white columns) and (a, b black columns) at species (a) and generic (b) levels, based on analysis of 819 species and 483 genera.

fluorescence in situ hybridization

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 450 species and 302 genera of the families , , , , and .

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 387 species and 263 genera.

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 52 species and 29 genera.

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 50 species and 23 genera of the families , , , and .

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 10 species and 9 genera of the family .

Histogram showing the distribution of female diploid chromosome numbers in at species and generic levels, based on analysis of 511 species and 335 genera of the families , , , , , , , , , , and .

Since the Halkka’s (1959) excellent review, the comparative cytogenetics of has never been rigorously addressed. The only exception is the two-part paper of Emeljanov and Kirillova (1990, 1992), which presents a comprehensive analysis of chromosome numbers and their variation at different taxonomic levels within every auchenorrhynchan family explored at that time. Thus, nearly fifty five years after Halkka’s and twenty five years after Emeljanov and Kirillova’s publications, we discuss here different aspects of cytogenetics of and summarize progress and problems in the field.

Chromosome structure

The overwhelming majority of eukaryotic organisms have monocentric chromosomes. These chromosomes possess the localized centromere, a region where two chromatids join and where spindle fibers attach during mitosis and meiosis. Like all , have holokinetic (holocentric) chromosomes. In contrast to monocentric chromosomes, holokinetic chromosomes have no localized centromere. The latter is considered to be diffuse and is formed by a large kinetochore plate (a circular plaque structure on the centromere by which the chromosomes are attached to spindle polar fibers) extending along all or most of the length of the holokinetic chromosome (Schrader 1947, Wolf 1996). Holokinetic chromosomes are sometimes designated as holocentric despite the fact that they lack a proper centromere. These chromosomes occur in certain scattered groups of plants and animals, being particularly widespread in insects, including (), (), , , (), , () (White 1973), and the enigmatic (Kuznetsova et al. 2002). Thus, holokinetic chromosomes occur in every major phylogenetic lineage (cohort) of suggesting that they are likely to have evolved at least four times independently in insect evolution.

In theory, the large kinetochore plate facilitates rapid karyotype evolution via occasional fusion/fission events. Firstly, fusion of holokinetic chromosomes would not create the problems characteristic of a dicentric chromosome in monocentric organPageBreakisms (i.e. displaying chromosomes with localized centromeres). Secondly, fission of a holokinetic chromosome should create chromosome fragments that exhibit a part of the kinetochore plate and can attach themselves to the spindle fibers at cell divisions. As a result, chromosome fragments that would be acentric (lacking a centromere) and hence lost in organisms with monocentric chromosomes may be inherited in holokinetic organisms. The gametes harboring chromosome fragments are consequently expected to be viable (Hipp et al. 2010). Fusion/fission rearrangements are therefore conventionally accepted as the commonest mechanisms of chromosome evolution in holokinetic groups. This assumption seems to receive support from the fact that the greatest range of within-genus variation in chromosome number related to the fusion/fission rearrangements is described in organisms with holokinetic chromosomes (reviewed in Kuznetsova et al. 2011). The evidence for the unique potential of holokinetic chromosomes’ fissions is provided by the blue butterfly Elwes (), 2n = ca 448-452, holding the record of the highest number of chromosomes in the non-polyploid eukaryotic organisms (Lukhtanov 2015).

Although variations in chromosome number of related species are probably due to both fissions and fusions, fusions are suggested to be more common in holokinetic groups (White 1973) including (Halkka 1959, 1964). The point is that a chromosome, whether holokinetic or monocentric, has to display two functional telomeres in order to survive a mitotic cycle. A chromosome resulting from a fusion event will always display two functional telomeres originated from the two ancestral chromosomes, whereas a chromosome from a fission event will have to be able to develop a functional telomere de novo (White 1973, Nokkala et al. 2007).

Chromosome numbers and possible trends of their evolution

Variation in chromosome number. The currently known diploid chromosome numbers in range between 8 and 38 (here and elsewhere chromosome numbers are provided for females), being the lowest in () and the highest in ( and ). The infraorders differ in the limits of variation in chromosome number and in the modal numbers (sometimes referred to as the type numbers or basic numbers). Within each infraorder, many taxa have more than one modal number and these are characteristically lower in than in . In , chromosome numbers vary from 2n = 8 ( sp. from ) to 2n = 32 ( Fallén from ). The numbers in most cicadomorphan species lie between 16 and 22, with rare exceptions above and below these limits. In , chromosome numbers vary from 2n = 20 ( Distant from ) to 2n = 38 ( spp. from and Sahlberg from ) with strongly marked modes at 28 (prevailing), 30 (second) and 26. The variation in chromosome number in various groups of is shown in Figs 1–8.

Despite the fact that all possess holokinetic chromosomes, many higher taxa of the suborder show stable or only slightly variable karyotypes. Quite often the chromosome number is constant within the same genus. Within , the PageBreakgenera Le Peletier & Serville, Ashmead, and Zetterstedt are examples. In the first, all three studied species, Signoret, Burmeister, and Peletier & Serville, share 2n = 22; in the second, PageBreakall four studied species, Erichson, Walker, Signoret, and Signoret, possess likewise 2n = 22; in the third, four studied species, Boheman, Fabricius, Edwards, and Edwards, possess 2n = 16 (de Lello et al. 1982, Kirillova 1987). In the family , the large genus Germar is characterized by 2n = 30, whereas in the , the majority of species in the genera Amyot & Serville and Stål shows 2n = 20 and 2n = 22, respectively (Kirillova 1987). The most impressive examples of chromosome stability come from groups which have been extensively studied. Thus, nearly all species and genera of the subfamily () have 2n = 22 (Whitten 1965), while those (33 species) of the tribe () – 2n = 28 (Maryańska-Nadachowska et al. 2006, Kuznetsova et al. 2010). Similarly, within the family , almost all so far studied representatives of the tribe (9 species) have 2n = 30; those of the tribe (8 species) 2n = 28, while those of the tribe (16 species) – 2n = 26 (Kuznetsova 1986, Kuznetsova et al. 2009a). The conservative numbers suggest no evidence that fusions/fissions have played a role in speciation and evolution of these groups.

By contrast, there are some groups in which a wide variety of chromosome numbers occurs suggesting that both fusions and fissions have established themselves during their evolution. In , within the genus Haupt, the 19 studied species vary broadly in chromosome number: 2n = 12, 14, 16, 18, and 20 (Halkka 1957, Juan 2011). The genus Walsh is another group, which seems to show a striking range in chromosome number. In this genus, the twelve species examined so PageBreakfar display 2n = 16 (4 species), 18 (2 species), 20 (4 species), and 22 (2 species) (Kirillova 1988, Aguin-Pombo et al. 2006, Juan 2011). This cosmopolitan genus with more than 1,000 described species is by far the most speciose genus in . is recognized as a genus requiring comprehensive revision (Southern and Dietrich 2010), and a cytogenetic approach might be useful to clarify the species-level systematPageBreakics of this group. Likewise, the eight species recognized in the remarkably polymorphic spittlebug genus Stål () display three different chromosome numbers in males: 20, 23, and 24 (Maryańska-Nadachowska et al. 2012).

The modal and ancestral chromosome numbers. Emeljanov and Kirillova (1992) argued that the ancestral chromosome numbers were as follows: 2n = 30 in , 2n = 26-28 in , 2n = 20 in , and 2n = 22 in . Although 2n = 22, discovered later in the most primitive membracoid family (Kuznetsova and Kirillova 1993), seemed to confirm the ancestrality of this number in , a more definite solution of this problem will be possible only after a more thorough investigation of every superfamily. For example, the currently available data on (308 species and 148 genera) are greatly skewed owing to the focus on the families and and on the tribe (), in which altogether over 160 species in 122 genera have been karyotyped. In , chromosome numbers are known for 50 species (23 genera), half of which belong to the family , while in only ten species (nine genera) in the have been studied. Within , 450 species in 302 genera have been karyotyped, of which at least 65% of species and 53% of genera belong to the subfamilies and (). It should be added here that Nielson, the recently studied first representative of the cicadomorphan superfamily (), was found to exhibit 2n = 18(16 + XY) (Golub et al. 2014). This chromosome number fits well into the range of most characteristic numbers in as a whole (16-22) being close to the numbers accepted by Emeljanov and Kirillova (1992) as putative ancestral ones for (2n = 20) and (2n = 22) but not for (2n = 26-28).

Opinions on the ancestral chromosome number in as a whole differ considerably (Halkka 1959, Kuznetsova et al. 1998, Emeljanov and Kirillova 1992). The solution of the problem very much depends on phylogeny accepted and method of ancestral number inference adopted. One approach to inferring the ancestral karyotype is mapping chromosome numbers typical for particular superfamilies on phylogenetic trees. Some researchers treat () as the most basal branch within (e.g. Shcherbakov 1984). This idea recently supported by molecular data (Cryan and Urban 2012) is tempting to speculate that the typical number for , 2n = 30 (Halkka 1959) or 2n = 28 (Kuznetsova et al. 1998), is the ancestral state in . However, based on a large number of morphological characters, Emeljanov (1987) hypothesized that the common ancestor of the four (now five including ) recent superfamilies differentiated first into cercopoid-cicadoid and fulgoroid-cicadelloid branches, thus becoming the clade + , the sister group of the clade + . Comparison of the putative ancestral numbers of the superfamilies (see above) with the Emeljanov’s phylogenetic scheme of + suggests that the characteristic karyotype of , 2n = 28 or 26, would be the most likely ancestral chromosome number in as a whole (Emeljanov and Kirillova 1992). PageBreakAnother approach to inferring the ancestral karyotype is a comparison of typical chromosome numbers between the target group and outgroups. Emeljanov and Kirillova (1992) used this approach and compared chromosome numbers in with those in other hemipteran lineages – , , , , and . This led to the conclusion that the common ancestor of all had a diploid set of 20-22 chromosomes.

Since Emeljanov and Kirillova’s (1992) publication, a large body of new cytogenetic data on hemipterans has been obtained. There is a good reason to reconsider the problem based on the data available at present time. In , diploid numbers vary from 4 to 192, with comparatively low numbers in “archaeococcoids” comprising the most basal families of scale insects – (2n = 14, 16, 18), (modal number of 2n = 4), and (2n = 8 in the only studied species Cockerell) (Gavrilov 2007). In , chromosome numbers vary between 4 and 72, the ancestral number most likely being between 8 and 20 (Blackman 1980) though, in our opinion, more likely between 8 and 22. In the most primitive families, 2n = 22 (), 2n = 8 and 12 () and 2n = 20 ( = ) seem to be most characteristic (Kuznetsova and Shaposhnikov 1973, Blackman 1980, Gavrilov-Zimin et al. 2015), but the sampling is still inadequate, at least for and (Gavrilov-Zimin et al. 2015). Although it is not necessary for a modal number to be ancestral in a group, it seems reasonable to assume that in the karyotype of 2n = 26, which has been conserved in 72% of the species and in 50% of the genera studied, is their ancestral trait (Maryańska-Nadachowska 2002). In , chromosome numbers are known for only four species (see Blackman and Cahill 1998): Westwood (2n = 22), Quaintance (2n = 26 and/or 28), Linnaeus (2n = 26), and Gennadius (2n = 20). In one of the most primitive true bug infraorder, the , chromosome numbers have been recorded for males of three representatives of the family : 2n = 20 + X(0) in Sahlberg, 2n = 20 + XY1Y2 in Sahlberg, 2n = 20 + XY in Hill, 2n = 20? + XY in Linnavuori, and for males of two species of the family : 2n = 32 + X(0) in Hill and Hill (Grozeva and Nokkala 1996). For the two studied species, Bergroth and Shcherbakov, male karyotypes of 2n = 26 + X(0) and 2n = 30 + X(0) respectively were recorded (Grozeva et al. 2014, Kuznetsova et al. 2015b).

In , a sister group to the rest of , the modal karyotype of 2n = 18 is considered as the ancestral one, although there appears to be considerable variation in chromosome number within more primitive suborder : 2n = 18 and 22 in , 2n = 20 in , and 2n = 30 in (Golub and Nokkala 2009). Thus, the available data on the higher hemipteran groups appear insufficient and too heterogeneous to reconstruct the chromosome number ancestral for .

Sex determining systems

Genetic sex determination predominates in higher animals, including insects, and is often accompanied by the presence of a heteromorphic chromosome pair in one sex (White 1973). The , in common with most other insects, display male heterogamety. The XX/X(0) sex determination (where 0 denotes the absence of the Y chromosome) is of common occurrence and seems to be an ancestral trait in this group (Halkka 1959, Emeljanov and Kirillova 1990, 1992) and in as a whole (Blackman 1995). Thus, in females two Xs are present, while in males only one X is present. Despite evolutionary stability, in some cases the X(0) system has been replaced by an XY system in species within the same genus that is otherwise exclusively X(0). When such cases occur in , it is often clear that the Y is a neo-Y (Blackman 1995). Examples of this are found in the genera Burmeister from (John and Claridge 1974) and from (Maryańska-Nadachowska et al. 2012). A highly peculiar situation occurs in the cytologically well-studied tribe (), in which all species have a neo-XY system and the only exception is Dufour with an X(0) system in males. In contrast to , species from all other tribes of this family are characterized by an X(0) system (Kuznetsova 1986, Kuznetsova et al. 2009a).

In organisms with XY systems, recombination between X and Y chromosomes is usually suppressed (White 1973) except for the cases when the XY is a neo-system. This type of sex determination usually arises from the ancestral X(0) system as a result of fusion between the original X chromosome and an autosome, the homologue of this autosome becoming a neo-Y chromosome (White 1973, Blackman 1995). Clearly, the derived karyotype should have one pair of autosomes less than the ancestral one. For example, the species of the tribe () have 2n = 26 + neo-XY, whereas those of the tribe 2n = 28 + X(0). In a recently formed neo-XY system, the autosomally derived Y chromosome (a neo-Y) and the autosomal part of the neo-X chromosome are still homologous, and therefore synapse in prophase I of meiosis. At metaphase I, the neo-XY bivalent is usually large and clearly heteromorphic, indicating a recent fusion between the X and an autosome pair.

Once a neo-XY system has arisen, it can undergo a further transformation into a multiple X1X2Y system as a result of a translocation involving the Y chromosome and another pair of autosomes. This may have occurred in the evolution of the sex chromosome mechanism in Drosopoulos & Remane, which has 2n = 20 + neo-X1X2Y against 2n = 22 + neo XY found in Melichar, Remane & Drosopoulos, and Drosopoulos & Remane (Maryańska-Nadachowska et al. 2012, 2013). Multiple sex chromosomes of the X1X2Y type, which form chiasmate trivalents in meiosis, have also been described in sp. (), however, in this case the X1X2Y system represents an example of sex chromosome polymorphism (Whitten 1968).

A different, achiasmate XY system, with a fairly small Y chromosome, is found in the planthoppers Oshanin and Kato () PageBreak(Kuznetsova 1986, Tian et al. 2004). The origin of the Y chromosome in these species is not entirely clear, but it seems likely that it has been derived from a mitotically stable B chromosome that has become a standard member of the karyotype, that is, the B chromosome transformed into the Y chromosome during evolution (discussed in section “Polymorphism for B-chromosomes”).

Polyploidy

Polyploidy, that is, multiplication of the chromosome set is well known to play a major role in speciation and evolution of plants, but is a fairly rare phenomenon in sexually reproducing animals (White 1973). Evolutionary polyploidy had occurred in a number of animal species that reproduce parthenogenically. In , all so far known cases of polyploidy are connected with parthenogenesis, either with gynogenesis, sometimes referred to “pseudogamy” (where the egg is activated by sperm borrowed from conspecific or closely related males, but without fusion of the egg and sperm nuclei), or with true parthenogenesis, more often referred to “thelytoky”. Although it is not necessary for all parthenogenetic forms to be polyploids, these are universally triploids in both leafhoppers and planthoppers. Only two planthopper genera, Wagner and Wagner (), are known to comprise a number of gynogenetic triploid forms (Drosopoulos 1976, 1977, den Bieman and Eggers-Schumacher 1986). In contrast, leafhoppers of the genus (Aguin-Pombo et al. 2006) and planthoppers of the genus Fieber (den Bieman and de Vrijer 1987) comprise triploid forms, which reproduce by true parthenogenesis. The genus is a good case in point. In this diverse, complex and cosmopolitan genus, the bisexual species are diploid, with hitherto known chromosome numbers of 2n = 16, 18, 20 and 22. In Madeira Island, besides the bisexual species Paoli (2n = 14 + XX), Ribaut (2n = 16 + XX) and DeLong (2n = 20 + XX), three all-female morphotypes (A, B and C) were discovered. In these females, the chromosomal complements are triploid, consisting thus of two female genomes and one male genome: 2n = 3x = 28 + XXX in morphotype A, 24 + XXX in morphotype B, and 21 + XXX in morphotype C. The study revealed that their reproduction follows an apomictic type (Aguin-Pombo et al. 2006).

In apomictic parthenogenesis, meiosis is completely suppressed, and eggs pass through a mitosis-like cell division, i.e. without formation of bivalents and recombination, and genetic heterozygosity is thus preserved. The heterozygosity is expected to be perpetuated from generation to generation, increasing slightly through mutations. It is generally proposed that most polyploid animals are allopolyploids, tending to be of hybrid origin (White 1973, Bullini 1985). It is assumed that such is most likely the case of the triploid forms of (Drosopoulos 1976). In contrast, in some groups, triploids seem to have an autopolyploid origin and this is probably true for triploids found in the genus (den Bieman 1988). The origin of the above-listed parthenoforms still remains unknown. Some of them seem to PageBreakbe closely related to bisexual species which are yet extant. Several hypotheses, including that of their hybrid origin, were made in Aguin-Pombo et al. (2006) but still much more information is needed to decide between the hypotheses.

Polymorphism for B-chromosomes

B-chromosomes (also referred to as supernumerary, additional or accessory) are chromosomes found in addition to chromosomes of the standard complement (A chromosomes) and occur in approximately 15% of living species (Beukeboom 1994). Little consensus has been achieved in understanding their origin, role, transmission, inheritance and evolution. B-chromosomes appear only in some individuals of some populations of the same species. Their presence is considered to be beneficial, harmful or neutral, and several authors consider B-chromosomes as parasitic and selfish (reviewed in Camacho 2004).

Small chromosomes additional to the standard complements and interpreted as B-chromosomes have been found in the leafhoppers Fallen and Boheman (Kuznetsova et al. 2013) and in several species of planthoppers (Halkka 1959, Booij 1982, den Bieman 1988, Kirillova and Kuznetsova 1990). Males of and were collected from a range of food plants in different localities in Greece. B-chromosomes were found in 3 out of 6 populations of and in 2 out of 7 populations of . A single B chromosome or sometimes two B-chromosomes were present in males. As is often the case, B-chromosomes were significantly smaller than chromosomes of the standard complements and negatively heteropycnotic during meiotic prophases and metaphases. No correlation was found between the occurrence and frequencies of B-chromosomes in populations with habitat or altitude (Kuznetsova et al. 2013).

It is suggested that inter-population differences in B chromosome distribution depend on selective factors (Camacho 2004). It is interesting to note in this connection that in the planthopper species Fabricius, Sahlberg, and Matsui (), B-chromosomes were present only in populations inhabiting Northeast Siberia and Kamchatka, whereas individuals sampled from different populations in European Russia (e.g. those of ) lacked B-chromosomes (Kirillova and Kuznetsova 1990). The point that should be mentioned is that both these planthopper species and aforesaid Fieber leafhoppers showed no more than 1 or 2 B-chromosomes per individual. One must suggest that this number is tolerable for B chromosome carriers and the natural selection operates to eliminate individuals with more than two B-chromosomes in all these populations.

The commonly accepted view is that B-chromosomes are derived from the standard complement of a species, including the X chromosome (Camacho 2004). On the other hand, the possibility of integrating a B chromosome into the standard chromosome complement has been suggested in a number of studies (reviewed in Nokkala et al. 2000). Recently, it has been claimed that the achiasmate Y chromosome in PageBreakFallén () might have evolved from a B chromosome (Carvalho 2002). To explain the formation of the achiasmate Y chromosome in separate species of psyllids (a related group sharing predominant X(0) sex determining system with ), Nokkala et al. (2000, 2003) suggested that the Y chromosome has evolved from a mitotically stable B chromosome that was integrated into an achiasmate segregation system with the X chromosome. Later, this chromosome would become fixed in the karyotype as the Y chromosome. In this connection, it should be noted that at metaphase I of some delphacid species, a B chromosome appeared closely associated with the univalent X chromosome forming a pseudo-bivalent XB (Kirillova and Kuznetsova 1990). The X and B-chromosomes tended to segregate (i.e. pass to opposite poles) at anaphase I suggesting thus an increasing transmission of B-chromosomes to sons while a decreased transmission to daughters. This means that the B-chromosomes are able to spread through the male line, whereas are removed from the female line. In this connection, it is interesting to note that a Y chromosome of unknown origin has been described in several species of the family (Kuznetsova 1986, Kuznetsova et al. 2010, see also section “Sex determining systems”). In these species, the XY chromosome pair was located outside the autosomal bivalents in the place where a univalent X chromosome is usually located in the X(0) auchenorrhynchan species. The XY pair differed distinctly from a neo-XY bivalent (characteristic, for example, of the closely related to family ) by its morphology and location at metaphase I. On the other hand, this XY pair appears to be identical to the XB pair described by Kirillova and Kuznetsova (1990) in , , and . According to Nokkala et al. (2003: 331), “the evolutionary dynamics of B-chromosomes, that is, the ability to transform into A chromosomes or vice versa, might have played a much more important role in the evolution of karyotypes than previously understood”.

Noteworthy is the different behaviour of B chromosomes in the leafhoppers and (Kuznetsova et al. 2013). In these species, B-chromosome(s) did not connect to the univalent X chromosome at MI, and when there were two B chromosomes in a set, they did not pair and passed randomly through meiosis as univalents, being still maintained in the populations by unknown means.

Other cases of polymorphism

Fission and fusion of holokinetic chromosomes do not result in unbalanced meiotic products, and so these rearrangements may be preserved through generations and establish variations in chromosome number within populations. Yet, descriptions of chromosomal polymorphisms are quite rare in . One can anticipate that it is due to very few studies at the population level in this group. However some chromosomal polymorphisms (other than polymorphism for B-chromosomes) do occur in natural populations of leafhoppers and planthoppers.

Polymorphism for sex chromosomes. Some cases of sex chromosome polymorphism were discovered in the leafhoppers sp. (Whitten 1968), PageBreak Distant (Manna and Bhattacharya 1973), Zetterstedt, and (Linnaeus) (John and Claridge 1974), as well as in the planthoppers Boheman () and Oshanin () (Halkka 1959, Kuznetsova 1986). In from India, the males were found to be dimorphic in sex chromosome constitution. Out of 30 males studied, 16 were of an X(0) type (designated as Type A), while 14 were of an XY type (designated as Type B). Type A was characterized by 17 chromosomes in the spermatogonial complement, with a pair of conspicuously large autosomes and a single medium-sized X chromosome. Type B displayed 2n = 18 with X being the largest, Y relatively small, and the pair of large autosomes present in Type A was absent. The sex chromosomes in Type B were suggested to have originated as neo-X and neo-Y by the X-autosome translocation from Type A (Manna and Bhattacharya 1973).

A very interesting example of sex chromosome polymorphism was revealed by John and Claridge (1974) in British populations of . In this species, mountain populations occurring on Ehrhart were X(0)-monomorphic, whereas populations in lowland woodlands were polymorphic, containing a mixture of X(0) and neo-XY males in the same or different populations.

Halkka (1959) studied a Finnish population of and found that some males in this population displayed a Y chromosome, while others did not. He suggested that the absence of Y chromosome in these males was a result of its loss, the Y being still inherited in part of the population.

Polymorphism for autosomes. Some impressive cases of a fission/fusion polymorphism for autosomes have been described in the Australian leafhopper species Kirkaldy (Whitten 1965) and Evans (Whitten and Taylor 1968), as well as in Greek populations of and (Kuznetsova et al. 2013). In eight studied populations of , specimens appeared invariable in having one of the following four chromosome complements: (1) three bivalents + X, (2) four bivalents + X, (3) two bivalents + one trivalent + X, and (4) one bivalent + one tetravalent + X. The reduction in the number of chromosomes has reached different stages in different areas. At Lake Pedder, the chromosome number was almost fixed (2n = 7, three bivalents + X), while at Bruny Island, there occurred a cline in chromosome number decreasing from north to south. This cline was caused by differences in the frequency of chromosome fusions. was suggested to be under a process of speciation driven by the reorganization of chromosomes that is initiated in some local populations through the fixation of a particular chromosome rearrangement (Whitten and Taylor 1968).

The brown planthopper Stål () is the only auchenorrhynchan species studied cytogenetically both from natural populations and laboratory cultures. It is notable that natural populations of this species across a wide geographic range revealed almost no instances of chromosomal polymorphism (den Hollander 1982), whereas males from the stock cultures showed a great amount of polymorphism (Liquido 1986, Goh et al. 1992). These differences in the level of chromosomal polymorphism between natural populations and laboratory cultures deserve further investigation.

Meiosis in males and females

Meiosis in normal spermatogenesis. Within , some very interesting and highly aberrant chromosome cycles and anomalous types of meiosis occur in aphids, scale insects, whiteflies, and true bugs, including moss bugs () (White 1973, Blackman and Hales 1986, Normark 2003, Papeschi and Bressa 2006, Kuznetsova et al. 2011, 2015b). In contrast to all these insects, meiosis in is essentially simple and uniform in different species and follows the classical “pre-reductional” scheme: during first meiotic division homologous chromosomes undergo pairing, synapsis and recombination at prophase I and segregation at anaphase I. As with autosomes, sex chromosomes undergo pre-reductional meiosis. During second division, sister chromatids separate and migrate to opposite poles at anaphase II creating then haploid daughter cells.

The number of chiasmata in bivalents. It is common knowledge that in meiosis, chiasmata (presumed to be the points of genetic crossing-over) are formed uniting homologous chromosomes together until their separation in the reductional division. In most organisms there are one to three chiasmata per bivalent, although in some organisms the number of chiasmata in a bivalent (i.e. the chiasma frequency) varies considerably being typically higher in plants than in animals (White 1973). Halkka (1964) analyzed the number of chiasmata in males and females of species belonging to six families of , namely, , , , , , and . In all species, bivalents were found to display one or occasionally two chiasmata, and no great differences in chiasma frequencies were detected between males and females, as well as between leafhoppers and planthoppers.

Similarly, the low number of chiasmata (estimated to be 1-2 from cytogenetic analyses) is a rule in psyllids (Maryańska-Nadachowska 2002) and true bugs (Kuznetsova et al. 2011) and is suggested to represent one of the peculiar features of holokinetic bivalents as such (Nokkala et al. 2004). Based on a detailed analysis of meiosis in a psyllid species, Flor, Nokkala et al. (2004) concluded that the cells carrying more than two chiasmata were inevitably eliminated creating thus a strong selection against the formation of multiple chiasmata in holokinetic bivalents. The main cause of this pattern was suggested to be a specific condensation process inherent to holokinetic chromosomes. It is worth noting that bivalents with multiple chiasmata have been observed occasionally in holokinetic groups, including (see for references Kuznetsova et al. 2009b); however, these observations never advanced beyond metaphase I of spermatogenesis, and therefore, the further fate of the cells with multichiasmate bivalents remained unknown.

Meiosis in normal oogenesis. In comparison to the rather abundant data available on male meiosis in , there have been no comprehensive investigations of chromosome behaviour in female meiosis. The only exceptions are the few descriptions of meiosis in parthenogenetic forms (see section “Polyploidy”) and the studies done by Halkka (1959, 1964) on female meiosis of several bisexual species. First, Halkka (1964) provided evidence for the low chiasma frequencies in females. PageBreakSecond, he revealed that females of the leafhopper species Metcalf displayed pre-reductional meiosis both for autosomes and sex chromosomes and one chiasma per bivalent (in contrast to two chiasmata in males) (Halkka 1959).

Meiotic abnormalities. It is to be noted that the apparent uniformity of meiosis in could be due to the small number of species which have been studied in any detail. The incidence of meiotic abnormalities and their relationship with different spermatogenic parameters was assessed in the leafhopper species and (Kuznetsova et al. 2013). Several isolated populations of these species in Greece sampled from different food plants, such as Mill., L., L., Mill., and spp., showed a great deal of meiotic abnormalities in males, including end-to-end non-homologous chromosomal associations, heterozygous translocation chains, univalents, anaphasic laggards, and aberrant sperms. The primary causes of abnormal chromosome behavior in studied populations, whether those are male-specific meiotic mutations or some environmental mutagens, remained unknown. Also it is not known whether these meiotic abnormalities may play a role in the genome diversity and karyotype evolution of the genus . The resolution of the issues will have to await further studies (Kuznetsova et al. 2013).

New approaches to cytogenetic studies

White (1978) estimated that over 90% of all speciation events are accompanied by karyotypic changes. Current evidence shows that chromosome numbers in are quite often remarkably conservative within a group despite the theoretical capacity of holokinetic chromosomes for fusion and fragmentation. It is to be noted, however, that, in general, cytogenetic studies of use standard techniques, providing evidence for chromosome numbers, sex chromosome mechanisms, and, in outline, the behaviour of chromosomes during meiosis. Nevertheless, for a student of auchenorrhynchan cytogenetics (as well as for an investigator of any other holokinetic group), the main challenge is the identification of individual chromosomes and chromosomal regions. This information would result in considerable progress in the field because it will allow identification of the interchromosomal and, what is more important, intrachromosomal rearrangements involved in the evolution of holokinetic organisms.

Chromosome banding is a staining technique to reveal differentiation within chromosomes as a series of reproducible cross-bands. Besides the identification of individual chromosomes in a karyotype, the bands tell a good deal about fundamental aspects of the chromatin organization and compartmentalization of the genome. These techniques have had an invaluable impact on plant and animal cytogenetics but still are very little used in . In this group, a number of studies have applied some conventional techniques, such as C-banding, AgNOR-banding, and DNA base specific fluorochrome-banding. C-banding characteristically reveals the extent and location of heterochromatic segments (C-bands), which contain highly PageBreakcondensed, repetitive and largely transcriptionally silent DNA. Fluorochrome-banding mainly involves GC-specific antibiotic chromomycin A3 (CMA3) and AT- specific 4-6-diamidino-2-phenylindole (DAPI) to detect variation in base composition along the chromosomes. AgNOR-banding reveals the nucleolus organizer regions (NORs), containing the genes that code for ribosomal RNA. These techniques have proved their utility for comparative purposes at the generic level. For example, the C-banding technique showed that taxonomically related species sharing the same chromosome number differ often in chromosome constitution due basically to the accumulation of many rearrangements since divergence from the common ancestor. For instance, differences in C-banding pattern were described between the delphacids and Horváth (Noda and Tatewaki 1990), between the cicadas Motschulski and Matsumura (Perepelov et al. 2002), between the species of the spittlebug genus (Kuznetsova et al. 2003, Maryańska-Nadachowska et al. 2008, 2012), and between the species of the family (Kuznetsova et al. 2009b, 2010). The issid species Dlabola and Fieber, both with 2n = 26 + X(0) in males, were shown to differ considerably in the amount of C-heterochromatin, which appeared clearly more abundant in the first of these species. The species differed also in C-heterochromatin distribution along the karyotypes and its ability to stain with DAPI and CMA3 (Kuznetsova et al. 2009b). On the other hand, the application of AgNOR-banding showed that and were similar in having NORs located sub-terminally in the largest pair of autosomes. It should be noted, that such a location of NORs seems to represent the most common pattern in as a whole (Kuznetsova et al. 2003, 2009b, 2010, Maryańska-Nadachowska et al. 2006, 2012).

In the last few decades, the ability to identify individual chromosomes in a karyotype has been markedly improved by the development of molecular cytogenetic techniques. These include, for example, fluorescence in situ hybridization (FISH) to locate the positions of different genes and specific DNA sequences on chromosomes, for analyses of genome homology, to identify alien chromosomes or segments, and immunofluorescence to detect the location and relative abundance of the proteins. Some of these techniques have been applied to economically important holokinetic species (Mandrioli et al. 2003, Mandrioli and Borsatti 2007, Marec et al. 2010, Grozeva et al. 2010, 2015), but have not yet been developed specifically for . The only exceptions are the Southern hybridization of genomic DNA with a telomeric probe and FISH of chromosomes with telomeric and ribosomal probes, which have been applied successfully to several auchenorrhynchan species (Frydrychová et al. 2004, Maryańska-Nadachowska et al. 2013, Golub et al. 2014, Kuznetsova et al. 2015a).

comparative genomic hybridization

genomic in situ hybridization

Telomeres are defined as the regions of the chromosomal ends that are required for complete replication, meiotic pairing, and stability of a chromosome (Zakian 2012). The molecular structure of telomeres is characterized by a tandem repeat of a short DNA sequence that is diversely differentiated in eukaryotes. Comparative PageBreakanalysis of these repeats (motifs) in various groups of organisms showed that they are evolutionarily stable, and, having once appeared during evolution, define taxa and phylogenetic branches of high rank (Traut et al. 2007). Frydrychová et al. (2004), using the Southern Hybridization technique, demonstrated the presence of telomeric TTAGG sequences in the genome of (). However, this technique is capable to reveal a sequence but not its chromosomal location within a genome. In contrast, in situ hybridization (FISH) is a technique that allows precise localization of specific DNA sequences on chromosomes.

Maryańska-Nadachowska et al. (2013) and Golub et al. (2014) pioneered in applying FISH to . In situ hybridization with the telomeric (TTAGG) and 18S rRNA gene probes was used to study eight species of the genus , (Maryańska-Nadachowska et al. 2013) and , (Golub et al. 2014). In most eukaryotic genomes, ribosomal DNA (rDNA) consists of tandemly repeated arrays of three genes (18S, 5.8S, and 28S) encoding nuclear rRNA and separated by internal spacers (Hillis and Dixon 1991). These arrays make up the and can be found clustered in one or several regions of the genome. First, the telomeric repeat probe confirmed that the chromosome ends of spp and are composed of the (TTAGG) nucleotide sequence, a common motif of insect telomeres. This motif was reported in the vast majority of evolutionary lineages in and is suggested to represent an ancestral sequence of telomeres in insects (Sahara et al. 1999, Frydrychova et al. 2004, Lukhtanov and Kuznetsova 2010). Second, the 18S rRNA gene probe showed that in 18S rDNA loci are placed on a medium-sized pair autosome and that species differ from one another in both number and location of major ribosomal gene loci in their karyotypes. Thus, the application of in situ hybridization technique to species showed an extensive reorganization of their genomes: the ribosomal genes changed repeatedly their relative position along the chromosomes indicating that a large number of rearrangements probably occurred during or soon after the species formation.

nucleolus organizing regions

Evolutionary relationships revealed from chromosome data

Given that chromosomes represent morphology at small scale, they can be used in phylogenetics in the same way as other morphological characters and can contribute to clarifying the systematics and phylogeny of a particular group.

Chromosome data have contributed to establishing the evolutionary relationships in several different ways which, except for rare occasions (e.g. Blackman 1980, Emeljanov and Kirillova 1990, 1992, Gokhman 2006, Angus and Tatton 2011; for some interesting references see also Lukhtanov and Kuznetsova 2010), are usually ignored by entomologists. As in other insect groups, cytogenetic data have been applied at both higher and lower rank levels in taxonomic and/or phylogenetic studies in .

Evolution at and above family level. Cytogenetic data placed in a phylogenetic context can provide insights into chromosome evolution within a higher rank taxon. PageBreakA number of successful examples of this approach have been made in (e.g., Emeljanov and Kirillova 1990, 1992, Kuznetsova et al. 2009a, Maryańska-Nadachowska et al. 2012). One of those deals with the planthopper subfamily (). This group comprises 192 species in 37 genera of four tribes: the Palaearctic ones (7 genera, 43 species), (1 genus, 1 species), and (20 genera, 104 species), and the Nearctic tribe , with 37 species in 10 genera (Emeljanov 1980, Emeljanov et al. 2005). For construction and substantiation of the phylogeny of , three types of characters were used: morphological (Emeljanov 1980), incomplete cytogenetic (Kuznetsova 1986, Kuznetsova et al. 2009a), and preliminary molecular (Emeljanov et al. 2005). As a whole, chromosome complements of 30 species (more than 15% of the total species recognized in ), belonging to 17 genera (almost one-half of the total) and all tribes, except for the African monotypic tribe , are currently known. Chromosome numbers being combined with some anatomical data (testis structure in terms of the number of testicular follicles) provided a strong support for the monophyly of and the recognition of two tribes, and . All were shown to have 26 autosomes, a simple sex chromosome mechanism of an X(0) type (2n = 26 + X), and male testes each composed of 6 follicles. On the other hand, species belonging to were described as having a pair of autosomes less, a secondary neo-XY sex determining system (2n = 24 + neo-XY), and 4 follicles per male testis. The karyotype of was suggested to have evolved by the fusion of two autosome pairs in an ancestral karyotype of 2n = 28 + X(0) (inherent to the second subfamily ). The karyotype of , in its turn, had originated from that of by an X-autosome fusion. Besides the gradual reduction of the total number of autosome pairs, there is apparently a trend towards the reduction of the number of testicular follicles in the evolution of . Thus, karyotype and testis structure both suggest a “basal” position of within . The states of these characters encountered in (2n = 24 + neo-XY and 4 follicles) are treated as being derived from those of (2n = 26 + X and 6 follicles), which is in good agreement with the morphological data (Emeljanov 1980). The point of interest is the tribe . The latter shows a number of morphological apomorphies and, therefore, is considered to be one of the most advanced tribes within (Emeljanov 1980, Emeljanov et al. 2005). Despite of this, three recently studied species of ( Stål, Ball & Hartzell, and Ball) were found to share the same karyotype and testis structure as the basal tribe (Kuznetsova et al. 2009a).

Additional examples showing the significance of chromosome data for the systematics and phylogenetics of are given below. As noted above, differ distinctly from in chromosome numbers (Fig. 8). The planthopper families and are also a good case in point. Emeljanov (1979) identified a number of important morphological differences that support the discreteness of these families. and are also distinguished by the morphology and the assumed origin of their Y chromosomes PageBreak(Kuznetsova et al. 2009a; see also sections “Sex determining systems” and “Polymorphism for B-chromosomes”). Opinions vary broadly on the phylogenetic position of the planthopper family . Although this family is mainly accepted as the most basal one within , some morphological and molecular evidence suggest that it is a relatively derived lineage among fulgoromorphans (see for references Yeh et al. 2005 and Urban and Cryan 2007). The currently available chromosome data seem to be consistent with this opinion: all hitherto studied species of display the evolutionarily derived neo-XY sex chromosome system (Kuznetsova and Kirillova 1990, Kirillova 1993).

Evolution below family level. Cytogenetic data also provide useful information about lower rank taxonomic relationships. For example, in the leafhopper Herrich-Schäffer, Halkka (1959) discovered two types of populations in Finland, i.e. with 2n = 14 + X and with 12 + XY in males respectively. He treated these differences in terms of a common polymorphism for an X-autosome translocation resulting in the formation of a neo-XY system. However, shortly afterwards, Vilbaste (1962) showed that the putative “12 + XY race” of was in fact Sahlberg.

Whitten (1965) described variation in chromosome number in different populations of a leafhopper species in Australia. An examination of male genitalia by J.W. Evans revealed that two morphologically distinct species were present. The 2n=11 group belonged to a species which, subsequent to Whitten’s (1965) paper, was separated from the genus as Kirkaldy, while the remaining chromosomal groups were morphologically uniform and all included in the new species, (Evans 1966).

The meadow spittlebug genus () is likewise a good example. This genus has been studied using morphological (Drosopoulos and Remane 2000, Drosopoulos 2003), molecular (Maryańska-Nadachowska et al. 2010, Seabra et al. 2010) and cytogenetic (Kuznetsova et al. 2003, 2015a, Maryańska-Nadachowska et al. 2008, 2012, 2013) techniques. Numerous studies have explored the outstanding colour polymorphism and systematics of this genus (e.g. Drosopoulos and Remane 2000, Drosopoulos 2003). A total of eight species are presently recognized, including the Mediterranean species Melichar, , Abdul-Nur & Lahoud, Melichar, Drosopoulos & Remane, Remane & Drosopoulos, and Drosopoulos & Remane, and the Holarctic species Linnaeus (Drosopoulos and Remane 2000). Based on morphology, the genus is currently divided into two groups: the “ species group and the “ species group (Drosopoulos and Remane 2000), whereas based on larval food plant preferences, the genus is divided into three ecological groups: developing on the lily Brot., on xerophilic plants, and on various dicotyledonous and monocotyledonous plants (Drosopoulos 2003).

The results of a recent phylogenetic study of using nucleotide sequences from two mitochondrial (COI and CytB) genes and one nuclear (ITS2) region are in general agreement both with the morphological and the food plant preferences PageBreakgroupings (Maryańska-Nadachowska et al. 2010). Likewise, differences in karyotype were found to be largely in agreement with the recognized groupings proposed on the basis of morphology and on the basis of larval food plant relationships. Cytogenetic analysis has revealed that a number of species share the same karyotype while some others differ in chromosome number, sex chromosome system and additional cytogenetic characters. The species feeding on were shown to have 2n = 22 + neo-XY (, and ) or 2n = 20 + neo-neo-X1X2Y (). These species are included into the species group “. Among the species of the “ group, and , with larvae developing on arid plants, share 2n = 18 + neo-XY, whereas and the polyphagous species , feeding on a wide range of dicotyledonous plants, possess 2n = 22 + X(0). It has been postulated that the ancestral karyotype of is 2n = 24 + X(0) and that karyotype changes occurred several times independently in the genus (Maryańska-Nadachowska et al. 2012, 2013).

In conclusion, it may be said that one of the most important ways of increasing the taxonomic and phylogenetic inferences based on chromosome data is to enlarge sampling of taxa. Considerable progress in our understanding of the cytogenetics of will come from the development and application of new molecular cytogenetic techniques, which appear clearly advantageous for revealing important markers in holokinetic chromosomes. These techniques are expected to provide useful insights into the genome constitution and mechanisms of karyotype evolution in this large group of .