Evolution of chromosome number in grasshoppers (Orthoptera: Caelifera: Acrididae)

Orthoptera have some of the largest genomes of all insects. At the same time, the architecture of their genomes remains poorly understood. Comparative cytological data across a wide range of taxa, even for basic parameters such as chromosome number, may provide important insights into the evolution of these genomes and help answer the question of why some species attained such large genome sizes. We collected and compiled more than 1,000 records of chromosome numbers of 339 genera (13.8% of 2,452 known genera) and 769 species (6.2% of 12,250 known species) of Caelifera, the suborder of Orthoptera that includes those taxa with short antennae. Within the family Acrididae, most of the records come from the subfamilies Oedipodinae (N = 325), Melanoplinae (N = 192) and Gomphocerinae (N = 254). Out of the 621 investigated species of Acrididae, 459 (73.9%) shared a chromosome number of 2n♂ = 23. Chromosome numbers of 2n♂ = 17 (12.2%) and 2n♂ = 21 (9.9%) were less common. The remaining 4.0% of species exhibited different chromosome numbers between 2n♂ = 8 (6 + XY) and 2n♂ = 27. Plotted on a phylogenetic tree, our results confirm that chromosome numbers, especially in the largest grasshopper family Acrididae, are highly conserved with a basic count of 2n♂ = 23 (22 + X0), sometimes reduced to, e.g., 2n♂ = 17 (16 + X0) in some genera of the slant-faced grasshopper subfamily Gomphocerinae. Species with divergent chromosome numbers occur in many of the groups we studied, but are not a systematic trait and have evolved multiple times independently. Our study supports the view that chromosome numbers are much more stable across the investigated Caelifera compared to Ensifera, the second suborder of Orthoptera that includes the long antennae bush crickets and crickets. Our results significantly extend our knowledge on the diversity of this character in Caelifera.


Introduction
Prior to the genomic era, cytogenetic studies provided the foundation for our understanding of animal genome organization (Bugrov, 1988(Bugrov, , 1996Bugrov & Vysotskaya, 1981;Confalonieri et al., 1998;Gokhman & Kuznetsova, 2006;King, 1995;Kirkpatrick, 2010;Vandergast et al., 2017). While genetic and genomic sequencing have become far more popular fields of research, cytogenetic studies still provide important information about the genomic organization of a species and clues to the evolution of whole groups of taxa (White, 1973). They have been used to address a variety of systematic, evolutionary and phylogenetic questions in plants and animals and have helped to improve our understanding of speciation (Charlesworth, 2004;Charlesworth & Charlesworth, 2005;Grzywacz et al., 2019;Navarro & Barton, 2003). Comparative cytogenetics Martin Husemann and Lara-Sophie Dey contributed equally to this work implements relatively simple studies of chromosome numbers and morphology, but it may also include more complex analyses of various banding patterns or highly specified gene probes with fluorescent staining (White & Solt, 1978;Zhong et al., 1996;Gokhman & Kuznetsova, 2006;Bishop, 2010). While these complex methods allow fine-scale analyses on the level of populations, comparative studies of chromosome numbers may give us insight into the higher levels of evolutionary processes.
Grasshoppers of the family Acrididae (Orthoptera: Caelifera) have been the target of intense cytogenetic studies (Cigliano et al., 2021). This group has been suggested to be relatively uniform in their chromosome number, with some exceptions (Hewitt, 1979;John & Hewitt, 1966). While a diploid chromosome number of 2n♂ = 23 (22 + X0) is considered the basic plan for Acrididae (Hewitt, 1979), different kinds of rearrangements, especially Robertsonian fusions, led to a reduction in chromosome number in some groups of Caelifera (e.g., many Eurasian Gomphocerinae have 2n♂ = 17 (16 + X0) chromosomes). McClung (1917) considered this variation in the number of chromosomes to be a matter of rearrangements of chromatin rather than a result of the loss or gain of individual chromosomes. Besides this, some variation in the sex determining system has led to variation in chromosome number. In general, loss of the Y chromosome led to the highly conserved sex chromosome pattern of X0♂/XX♀ found in most species. Due to several chromosome rearrangements (autosomes and sex chromosomes), some species evolved several alternative sex determining systems, e.g., neo-XY♂/XX♀ or even neo-X 1 X 2 Y♂/ neo-X 1 X 1 X 2 X 2 ♀ or X 1 X 2 0♂/X 1 X 1 X 2 X 2 ♀ (Palacios- Gimenez et al., 2013Gimenez et al., , 2018Castillo et al., 2010;Hewitt, 1979;White, 1973) leading to some variation in chromosome number and providing a possible basis for reproductive isolation in some species groups.
Despite some exceptions, in comparison with its sister group, Ensifera (katydids, crickets and allies), the variation in chromosome number is relatively lower in Caelifera. Also, in general, the chromosome number appears to be lower in Caelifera compared to most Ensifera, as Warchałowska-Śliwa (1998) reported a basic number of 2n♂ = 31 (30 + XO) chromosomes in males of most of the investigated subfamilies of the family Tettigoniidae. Interestingly, genome sizes are, regardless of the chromosome number, much smaller in Ensifera compared to Caelifera, which may suggest some duplication events at the advent of the diversification of Caelifera (Husemann et al., 2021;Mao et al., 2020). A recent meta-analysis of Polyneoptera showed that many interacting factors underlie chromosome variation (Sylvester et al., 2020). Warchałowska-Śliwa (1998) summarized the cytogenetic information of about 400 species of Tettigoniidae with the aim of tracing the evolution of chromosome number in that ensiferan family. Such a systematic review and analysis of chromosome number and evolution is lacking for the diverse caeliferan family Acrididae. Hence, with the aim of closing this gap, we provide new karyotype data of 36 species (and additional estimates of 8 previously investigated species) of Acrididae and assembled a dataset of 1,284 records of chromosome numbers for Caelifera representing 339 genera (13.8% of 2,452 known genera) and 769 species (6.2% of 12,250 known species), including 1,108 records of Acrididae. We provide an overview of the variability of karyotypes for several subfamilies of Acrididae and map chromosome numbers on the most recent phylogeny of Caelifera (Song et al., 2018) in order to get an insight into the evolution of chromosome number in this diverse group of grasshoppers.

Material examined
We collected male grasshoppers belonging to 16 species of Oedipodinae by sweep net sampling on field trips between 2014 and 2016 (SI 1).Voucher specimens were deposited at the entomological collection of the Zoological Museum Hamburg, Germany (ZMH). David B. Weissman and David Lightfoot have collected and analyzed western US grasshoppers over the years and we included 28 unreported results herein (DBW, unpubl. Data).

Cytogenetic analyses
We dissected and fixed testes of the collected specimens in the field in a solution consisting of three parts of ethanolacetic acid (3:1, v/v). Specimens were fixed in 99.9% ethanol after dissection. The samples were subsequently stored in a freezer at -20 °C until further processing. NU conducted chromosome analysis: Testes were stained with an alcohol-carmine solution for several hours, before being transferred to glass slides for squash preparation, chromosome counts and microscopic imaging (see Lightfoot et al., 2011;Ueshima & Rentz, 1979). DBW material was analyzed as in Rentz & Weissman (1984).

Chromosome mapping and ancestral state reconstruction
We added our newly generated data to a large dataset based on previously published data: We screened the literature and additionally included all unique records from two online databases: www. bchrom. csic. es and www. coleo guy. github. io/ karyo types. In total, we gathered 1,284 records of chromosome numbers of Caelifera, including the 1,108 records of Acrididae used in our analyses. Throughout the manuscript, we only show the male chromosome numbers if the normal X0 sex determining system is realized. In cases of deviating sex chromosome configurations, these are noted.
We visualized the distribution of male chromosome numbers of Acrididae as a histogram in R using the packages ggplot2 (Wickham, 2016), ggpubr (Kassambara, 2020), scales (Wickham & Seidel, 2020) and cowplot (Wilke, 2019). All subfamilies were colored to display differences in numbers between taxa.
We mapped male chromosome numbers on the most recent phylogeny of the group (Song et al., 2018) using the R packages BiocManager (Morgan, 2019), phytools (Revell, 2012), vctrs , ggplot2 (Wickham, 2016), ggtree (Yu et al., 2018(Yu et al., , 2017, gtable (Wickham & Pedersen, 2019), grid (R Core Team, 2020), ggstance  and tidyverse . A parsimonybased ancestral state reconstruction of chromosome number was done in Mesquite v. 3.51 (The Mesquite Project Team., 2019) based on a character matrix approach. We chose this reconstruction method due to missing data of several groups within the tree and several chromosome number configurations occurring only single times throughout the evaluated taxa. Parsimony approaches for ancestral state reconstruction are known to be as accurate in state reconstruction of deep and shallow nodes as likelihood approaches (Holland et al., 2020). We performed all R analyses in R version 3.6.3 (R Core Team, 2020).

Cytogenetic analyses
All species newly analyzed here had a karyotype of 2n♂ = 23 chromosomes with no variation or heteromorphism in any of the specimens studied ( Fig. 1, SI 1) with the exception of Teicophrys californiae Descamps, 1977, which had 2n♂ = 17. All chromosomes were acrocentric or telocentric.

Mapping and ancestral state reconstruction
We mapped the chromosome number on the phylogeny provided by Song et al. (2018) (Fig. 2) and performed ancestral state reconstruction. In their phylogeny, Song et al. (2018) showed that Acrididae roughly form four monophyletic groups (Clades A to D). Our analysis shows that three of these groups (A to C) show different degrees of polymorphism in chromosome number, whereas the fourth group (Clade D) appears monomorphic with a consistent chromosome number of 2n♂ = 23. However, we were not able to obtain chromosome numbers for all taxa included in the phylogeny.
Clade C represents the most variable group within the dataset. The group contains several genera of the paraphyletic Ancestral state reconstruction (Fig. 2) suggests an ancestral chromosome number of 2n♂ = 23 for Acrididae. Changes in chromosome number across the phylogeny in most cases represent single species.

Discussion
Based on our dataset, we confirm a high stability of chromosome number in Acrididae with almost three quarters (73.9%) of all records reporting a number of 2n♂ = 23. This is in line with previous findings of White (1973), who suggested that two-thirds of all species have this karyotype, and findings of Aswathanarayana & Ashwath (2006) who even suggested that 90% of Acrididae share this configuration. Our study therefore confirmed the traditional view of Acrididae as a prime example of karyotypic conservatism (White, 1973), but provides a more comprehensive analysis.
Nevertheless, despite high degree of conservation of the chromosome number configuration of 2n♂ = 23, some groups exhibited deviations from this typical number: Several tribes of Gomphocerinae (i.e., Stenobothrini, Gomphocerini and Chrysochraontini) share a number of 2n♂ = 17, while several Tetriginae species show a configuration of 2n♂ = 13 (Bugrov, 1996). Many Pamphagidae genera show a general chromosome number of 2n♂ = 19 (e.g., Bugrov, 1986Bugrov, , 1996. Coleman (1948) suggested that this reduction in the chromosome number was the result of centric fusions, also known as Robertsonian translocations (Cabrero & Camacho, 1986). It is difficult to assess whether the event of chromosome number reduction occurred a single time or gradually in multiple events because the currently available phylogenetic data include only few taxa with this reduced chromosome number. As no intermediate forms have been found in any closely related groups, it may be possible that the reduction occurred in a single step as, for example, also suggested in Oxyopidae spiders (Stávale et al., 2011). However, intermediates may also be of meiotic disadvantage potentially explaining their absence.
A reduction to 2n♂ = 21 is fairly widespread in some genera of Melanoplinae, e.g., Hesperotettix and Dichroplus, and the Oedipodinae Trimerotropis and Circotettix. The North American tribe Trimerotropini was the subject of intense cytogenetic studies and showed variation in chromosome number between 2n♂ = 21 and 2n♂ = 23. Within this tribe, species of Trimerotropis and Circotettix show geographic variation in chromosome number, and several evolutionary scenarios have been developed, potentially explaining these differing chromosome numbers (Confalonieri et al., 1998;Confalonieri & Bidau, 1986;Evans, 1954;White, 1949;Coleman, 1948). White (1949) suggested that the ancestral state is the typical 2n♂ = 23 and proposed that the fusion of two acrocentric chromosomes to a metacentric chromosome has produced the decreased karyotype of 2n♂ = 21 in species of Circotettix and Trimerotropis. The metacentric chromosomes in other Trimerotropini genera originated probably by pericentric inversions (Evans, 1954), rather than translocations as suggested for the Gomphocerinae (Coleman, 1948). The effect of this chromosomal polymorphism for reproductive isolation remains debated: natural hybrids with 2n♂ = 22 have been observed in several crosses; yet, sperm quality strongly suffered in many cases suggesting some degree of hybrid sterility (Shaw et al., 1998;John et al., 1983;John & Weissman, 1977;Evans, 1954).
The two main chromosomal Sections A and B were also recovered in phylogenetic reconstructions using mitochondrial and nuclear genes (Husemann et al., 2012) and hence represent a useful systematic character. This still has to be evaluated in the Gomphocerinae.
In turn, some genera are particularly diverse in their chromosome constitutions, foremost the Melanoplinae genera Dichroplus and Hesperotettix. Chromosome numbers vary between 2n♂ = 18 (16 + XY) and 26 (24 + XY) within the genus Hesperotettix (McClung, 1917) and between 2n♂ = 8 (6 + XY) and 2n♂ = 27 in Dichroplus (Castillo et al., 2017;Mesa et al., 1982). Interestingly, there have been several studies performed on the chromosome number variation of the species Podisma sapporensis and Podisma pedestris in hybridization zones (e.g., Warchałowska-Śliwa et al., 2008;Bella et al., 1991). These studies show that reproductive isolation systems exist in hybrids, but the variation is most likely based on Robertsonian translocations between a sex chromosome and an autosome, and several chromosome rearrangements. Further, they show a clear differentiation into X0 and neo-XY chromosome races and complex chromosomal polymorphism in contact zones, which could permit the differentiation of several chromosomal races (Warchałowska-Śliwa et al., 2008).

Conclusion
Overall, a basic chromosome number of 2n♂ = 23 was observed across the whole Acrididae phylogeny and hence in all four clades described by Song et al. (2018). No subfamily with a number consistently diverging from the standard 2n♂ = 23 was recovered in the tree; but, some taxon-specific chromosome number variation appears to be present in Gomphocerinae and Trimerotropini. We conclude that the chromosome number in Caelifera, and specifically in Acrididae, is rather constant and phylogenetically less informative compared to several groups of Ensifera, which show more variation (e.g., Eneopterinae with range from 2n♂ = 9 (6 + XXY)  up to 2n♂ = 57 in Rhaphidophoridae (Vandergast et al., 2017 and references therein)). The reasons for this need to be further explored in the future.
Author contribution Martin Husemann took part in conceptualization and supervision; Martin Husemann, Lara-Sophie Dey, David Sadílek, Norihiro Ueshima and David Weissman were involved in methodology; Martin Husemann, Lara-Sophie Dey and Norihiro Ueshima carried out formal analysis and investigation, and acquired the funding and resources; Martin Husemann and Lara-Sophie Dey wrote and prepared the original draft; Martin Husemann, Lara-Sophie Dey, David Sadílek, Norihiro Ueshima, Oliver Hawlitschek, Hojun Song and David Weissman performed writing, review and editing; and Lara-Sophie Dey designed the layout.
Funding Open Access funding enabled and organized by Projekt DEAL. Orthopterists' Society grant to MH and LSD; Hintelmann prize to MH; personal funding of Heinrich Böll Stiftung to LSD.

Availability of data and material
The authors declare the availability of data in the supplementary section of the manuscript. Further material is available in the Zoological Museum Hamburg (ZMH).
Code availability Not applicable.

Declarations
Ethics approval Not applicable.

Consent to participate Not applicable.
Consent for publication All authors declare their consent to publication of the scientific article.

Conflicts of interest The authors declare no conflict of interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.