The trade‐off between investment in weapons and fertility is mediated through spermatogenesis in the leaf‐footed cactus bug Narnia femorata

Abstract Males have the ability to compete for fertilizations through both precopulatory and postcopulatory intrasexual competition. Precopulatory competition has selected for large weapons and other adaptations to maximize access to females and mating opportunities, while postcopulatory competition has resulted in ejaculate adaptations to maximize fertilization success. Negative associations between these strategies support the hypothesis that there is a trade‐off between success at pre‐ and postcopulatory mating success. Recently, this trade‐off has been demonstrated with experimental manipulation. Males of the leaf‐footed cactus bug Narnia femorata use hind limbs as the primary weapon in male–male competition. However, males can drop a hind limb to avoid entrapment. When this autotomy occurs during development, they invest instead in large testes. While evolutionary outcomes of the trade‐offs between pre‐ and postcopulatory strategies have been identified, less work has been done to identify proximate mechanisms by which the trade‐off might occur, perhaps because the systems in which the trade‐offs have been investigated are not ones that have the molecular tools required for exploring mechanism. Here, we applied knowledge from a related model species for which we have developmental knowledge and molecular tools, the milkweed bug Oncopeltus fasciatus, to investigate the proximate mechanism by which autotomized N. femorata males developed larger testes. Autotomized males had evidence of a higher rate of transit amplification divisions in the spermatogonia, which would result more spermatocytes and thus in greater sperm numbers. Identification of mechanisms underlying a trade‐off can help our understanding of the direction and constraints on evolutionary trajectories and thus the evolutionary potential under multiple forms of selection.


| INTRODUC TI ON
Male-male competition has led to directional selection on weapon size (Andersson, 1994;Emlen, 2008). Increasing weapon size can result in improved success in competition for territories and mating opportunities, but it can be costly. Extreme weapon size may be countered by the increasing costs through natural selection (O'Brien et al., 2017). This leads to a potential trade-off between investing in weapons and other important fitness traits, including investment in fertility and other postcopulatory traits (Filice & Long, 2018;Joseph et al., 2018;Lüpold et al., 2014;Somjee et al., 2018). Observational studies have demonstrated the predicted negative association between size of weapons and size of copulatory organs, including testes and ejaculates, supporting the idea of a trade-off between investment in traits leading to precopulatory and postcopulatory mating success. Males that lack weapons may compensate by investing more in any potential mating opportunity.
The trade-off between pre-and postcopulatory tactics has recently been experimentally tested in the leaf-footed cactus bug Narnia femorata Stål (Hemiptera: Coreidae). Males of N. femorata use their enlarged hind limbs to strike and squeeze other males over access to territories that attract females (Nolen et al., 2017;Procter et al., 2012). While hind limbs are crucial for winning fights with other males (Emberts et al., 2016), twelve percent of adult male N. femorata in the wild are missing one or more limbs through the process of autotomy (Emberts et al., 2016). Autotomy is used by N. femorata of all ages to escape entrapment, and autotomy does not reduce their survival in a laboratory setting (Joseph et al., 2018;Miller et al., 2021).
Autotomized limbs are not regenerated (Emberts et al., 2017;Emberts et al., 2018). Even though males missing a hind limb are poor competitors with other males, they may still have an opportunity to mate.
Females move around a lot, and males may occasionally encounter them alone (C. W. Miller, personal observation). Further, males may avoid other males by sneaking copulations (see Gross, 1996). Indeed, observations of N. femorata and other leaf-footed bugs from the wild and in seminatural enclosures have revealed that dominant males patrol territories with multiple females but cannot always keep other males away (C. W. Miller, personal observation).
Experimentally autotomized N. femorata males that have lost their weapon during development have reduced opportunity to invest in the weapons important for success in precopulatory competition.
Here, we examine the mechanism of these observed patterns, specifically whether the testes of autotomized male N. femorata show an increase in sperm production.
While much work has focused on the selection and evolutionary outcome of the trade-off between pre-and postcopulatory success, much less work has been done on the mechanisms by which males might respond to an environment in which that trade-off occurs. Yet, we know that both genetic and developmental processes can constrain or facilitate the evolution of traits (Smith et al., 1985).
Identifying these genetic and developmental mechanisms could be critical to understanding the targets of selection leading to the evolutionary trade-off (Zera & Harshman, 2001). Mechanisms may be understudied in part because the biological systems in which these trade-offs are being studied are not easily dissected using modern tools, or their development has not been deeply characterized. Narnia femorata, however, belongs to the Hemiptera, a diverse group of bugs that have been widely studied and for which genomic resources are being developed (Panfilio & Angelini, 2018). Here, we take advantage of extensive information on testis development and spermatogenesis in a relative of N. femorata, the large milkweed bug, Oncopeltus fasciatus, to begin to identify the developmental mechanism by which autotomized males may be increasing sperm production that results in larger testis size. While O. fasciatus does not display autotomy (to our knowledge), other variables have been shown to influence the progress through spermatogenesis (Duxbury et al., 2018).
O. fasciatus is a hemipteran bug with a sequenced genome (Panfilio et al., 2019) that has been used as a physiological and developmental model system. Testis development (Economopoulos & Gordon, 1971)

| Experimental bugs
The impacts of autotomy at the onset of the penultimate stage on spermatogenesis were investigated as hemipterans are known to experience dramatic testes growth in these later stages of juvenile development (Economopoulos & Gordon, 1971). Once nymphs reached the 4th (penultimate) instar, they were randomly divided into one of two treatments: induced autotomy of the left hind limb or no autotomy (baseline control). Autotomy was induced by grasping the left hind limb by the base close to the body with forceps, allowing the bug to pull away and create a break at the joint between the trochanter and femur (Emberts et al., 2016). Once the treatments were applied, the 4th instar nymphs were placed on cactus in groups of 3-4 siblings per cup as conspecific density can impact development. Within high concentrations of N. femorata, faster developing siblings have much larger body sizes compared with the last siblings to develop into adults, which suggests strong competition between siblings (Allen & Miller, 2020). By placing nymphs in groups of 3-4, the impacts of conspecific density on development were minimized and consistent for the groups. Within 48 hr of becoming adults, the male bugs were separated into their own individual cups with a cactus pad and fruit.

| Dissection and staining of testes
Between 21 and 28 days post-adult emergence, the testes were removed from males. At this age, males are sexually mature. In F I G U R E 1 Structure of Narnia femorata testis tubule. (a) A low-magnification image (6×) of a N. femorata testis tubule stained with DAPI (DNA). The progression through spermatogenesis could be clearly differentiated based on nuclear morphology. At the apical tip, spermatocysts containing spermatogonia were identified based on the number of nuclei and nuclear morphology. Spermatogonia (sg) have dense, uniformly stained nuclei. They undergo mitotic transit amplification divisions to give rise to spermatocysts containing 64 nuclei in Oncopeltus fasciatus (Ewen-Campen et al., 2013). Posterior to the spermatogonia undergoing transit amplification divisions, the spermatocytes (sc) undergo meiosis to form haploid spermatids (st), which then differentiate into spermatozoa (sz). (b) Higher magnification (12×) image of a representative testis tubule stained for both DNA (DAPI, cyan blue) and dividing cells (α-pHH3 antibody; magenta). In this image, within the apical Region 1 (demarcated with the yellow dotted line) 5 spermatocysts containing synchronously dividing spermatogonia were labeled (yellow arrows). Occasionally, single nuclei were labeled with antibody (yellow star). These are likely to represent endoduplication of the cyst cell nuclei and were not included in the counts as they were clearly not spermatogonia or spermatocytes given that all nuclei within a cyst divide synchronously. Posterior to Region 1, in Region 2 (demarcated with the white dotted line), spermatocysts at the boundary of the spermatocytes and spermatid are labeled with α-pHH3. While typically there are fewer spermatocysts dividing in this region, in this image there were 5 spermatocysts labeled as they progressed through meiosis, identified by the number of nuclei within the cyst (white arrows). Occasionally single cells were labeled (white star) in this region, again, presumably cyst cell nuclei O. fasciatus, most testis development occurs in juveniles, with some sperm maturation during sexual maturation (Economopoulos & Gordon, 1971). Once sexual maturation is reached, sperm production is at a "steady state" until mating. Thus, changes with age (days past adult emergence) in virgin males were not predicted for virgin sexually mature males. So for convenience, we grouped males at this stage as a single developmental stage, sexual maturation, rather than by age (days past adult emergence). Individual tubules were separated from the outer membrane for fixation and staining. Testis tubules from males within a treatment were pooled for staining if they were dissected on the same day. The testis tubules were fixed in 4% formaldehyde in PBS plus 0.1% Triton X-100 (PBT) for 30 min. The tubules were stained with α-phosphohistone H3 Ser10 (pHH3) primary antibody (Millipore antibody 06-570, Sigma-Aldrich). α-pHH3 stains for chromosome condensation in preparation for mitosis and meiosis (Hans & Dimitrov, 2001;Prigent & Dimitrov, 2003).
The secondary antibody was an Alexa Fluor goat anti-rabbit 647 (Thermo Fisher Scientific). Following antibody staining, the tubules were stained with DAPI (0.5 μg/mL PBT) to visualize nucleic acids.
Stained tubules were mounted in Mowiol 4-88 mounting medium (Sigma-Aldrich). Slides were kept in boxes to limit light exposure and stored at 4℃ until visualized. The testis tubules were visualized with a Zeiss LSM 710 Confocal Microscope (Zeiss) at the UGA Biomedical Microscopy Core Facility. All testis tubules were photographed and included in the analysis.

| Analysis of division rates within spermatogonia and spermatocytes
To test the prediction that autotomized males would show higher division rates, estimated from the number spermatocysts positively stained with α-pHH3, than intact males, the number of spermatocysts stained with the α-pHH3 antibody was scored in the photographs of individual testis tubules from males in the two treatments.
The images were divided into the two regions, and stained spermatocysts were counted separately for Region 1, containing spermatogonia undergoing mitotic transit amplification divisions, and Region 2, containing spermatocytes undergoing meiotic division (Figure 1b).
Only positively stained spermatocysts were scored. Results were reported on a single testis tubule basis. Single cells were occasionally stained, perhaps representing endoreplication within the cyst cells that enclose the spermatocysts (Figure 1b). These were not included in the analysis. Prior to analysis, the photographs were coded by an independent observer to allow for the data to be collected blind with respect to treatment. To check for interoperate error, two people counted a subset of images for stained spermatogonia in Region 1 and spermatocytes in Region 2. There was good agreement between the two sets of data indicating that the scoring was reliable. Values for positively stained spermatocysts ranged from 0 to 20 in fixed intervals. Visual inspection of the data indicated that the distributions met the assumptions of ANOVA. We analyzed 36 testis tubules for each treatment. We analyzed the data with a one-way ANOVA with fixed effect using JMP Pro version 14. Power was calculated within the model function of JMP Pro v.14 with α = 0.05.

| RE SULTS
Regions of spermatogenesis were easily identifiable within the testis tubule (Figure 1a)

| D ISCUSS I ON
While many studies have investigated the evolutionary outcomes of the trade-off among pre-and postcopulatory strategies, fewer have investigated the mechanisms by which that trade-off is mediated. We took advantage of the wealth of knowledge about a related model insect, O. fasciatus, to explore how loss of a weapon during testis development impacted spermatogenesis in N. femorata. We found an increase in testis size and fertilization advantage in autotomized male N. femorata that has been previously documented (Joseph et al., 2018;Miller et al., 2019;Miller et al., 2021) was associated with an increased rate of mitotic divisions in spermatogonia.

Plasticity in sperm numbers and quality under variable conditions
has been explored (Bunning et al., 2015;Dávila & Aron, 2017;Joseph et al., 2016;Moatt et al., 2014;Somjee et al., 2018), but few have examined the mechanism by which the increase in sperm numbers occurred. In Drosophila melanogaster, males respond to perceived sperm competition risk by increasing sperm production (Moatt et al., 2014). A recent study has shown that mating increases the division rate of germline stem cells in the testes of D. melanogaster through G protein signaling (Malpe et al., 2020). However, this sort of mechanistic studies depends on established molecular markers to identify the germline stem cells and niche cells, and genetic tools lacking in most species.
In O. fasciatus, and presumably N. femorata, sperm arise originally from germline stem cells at the tip of each testis tubule (Schmidt et al., 2002). As shown in D. melanogaster (Malpe et al., 2020), variation in sperm production could arise through the rate of production of germ cells through division of the germline stem cells. Alternatively, variation could arise through the modulation of the sperm production process (Extavour, 2013;Kaczmarczyk & Kopp, 2011;Moore, 2014;Ramm & Schärer, 2014). While we do not have the tools to examine germline stem cell turnover in either N. femorata or O. fasciatus, we have been able to show an impact of losing a weapon during larval development on the rate of spermatogonial divisions in adults.
The timing of autotomy in this study corresponded to a critical period of testis development. In O. fasciatus, testes are small and undeveloped in the 1st through 3rd instars. During the 4th instar stage of development, meiosis is initiated and spermatids begin to form (Economopoulos & Gordon, 1971;Schmidt et al., 2002). If, as we predict, the developmental timing is the same in N. femorata, then autotomy during the 4th instar stage of development is a time when the testis could benefit from increased resource allocation, either in increasing the number of germline stem cells (Kaczmarczyk & Kopp, 2011) or in increasing the number of spermatogonia that enter the maturation pipeline (Moore, 2014). The lack of difference in numbers of spermatocysts undergoing meiosis in Region 2 possibly could reflect the fact that these arose from spermatogonia born prior to autotomy.
The trade-off between weapons critical for precopulatory mating success and traits critical for postcopulatory fertilization success has now been demonstrated in a number of species. Males that lack weapons tend to have increased fertilization success (Joseph et al., 2018;Somjee et al., 2018;Van den Beuken et al., 2019). Males that invest in precopulatory traits may not be able to fully invest in postcopulatory traits (Parker et al., 2013;Parker & Pizzari, 2010;Simmons et al., 2017). The constraint on investment in these strategies could be genetic (Filice & Long, 2018) (Reznick, 1985;Reznick et al., 2000) or depend on resource allocation (Joseph et al., 2018;Somjee et al., 2018). While there is much to be discovered about molecular and physiological control of spermatogenesis in N. femorata, studies such as this will allow researchers to dissect prospective targets of selection at a molecular level, opening up the potential for more targeted experimental manipulation. Ultimately, integrating the fitness outcomes of these trade-offs with the molecular and cellular control mechanisms will allow us to examine the way in which selection shapes, or is constrained by, mechanism.

ACK N OWLED G M ENTS
The authors would like to acknowledge the financial support of the College of Agricultural and Environmental Sciences in support of KRC's undergraduate research. This study was funded by NSF grant IOS-1553100 to CWM.

CO N FLI C T O F I NTE R E S T
The authors declare there are no conflicts of interest with this work.

O PE N R E S E A RCH BA D G E S
This article has earned an Open Data Badge for making publicly available the digitally-shareable data necessary to reproduce the reported results. The data is available at https://datad ryad.org/stash/ datas et/doi:10.5061/dryad.5hqbz kh5m.

F I G U R E 2
The number of spermatocysts within a single testis tubule from control males and autotomized males of the leaf-footed cactus bug Narnia femorata that were positively stained with an α-pHH3 antibody. (a) Spermatocysts in Region 1 containing spermatogonia from autotomized males were stained more frequently than those in the testis tubules of control males. (b) There was no difference in the number of spermatocysts in Region 2 containing spermatocytes undergoing meiosis between autotomized and control males. Black dots indicated the mean value for each treatment, and gray dots are individual data points. Each error bar is constructed using 1 standard error from the mean

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets analyzed for this study are made available by the authors through the publicly available Dryad Digital Repository (https://datad ryad.org/stash/ datas et/doi:10.5061/dryad.5hqbz kh5m).