Density of sperm-producing tissue is positively linked to male reproductive success, but not to
testes size in the Japanese quail (Coturnix japonica)
Martine Graf
Degree project in biology, Master of science (2 years), 2020 Examensarbete i biologi 45 hp till masterexamen, 2020
Biology Education Centre, Uppsala University, and Centre for Ecology and Conservation at University of Exeter
Supervisors: Dr. Richard Svanbäck and Dr. Barbara Tschirren
External opponent: Julian Baur
1 | P a g e
Abstract
If sexual selection acts in a sex-limited way on a trait that has a shared genetic basis between males and females, the resulting intralocus sexual conflict can have a considerable impact on the opposite sex. A prime example of sexual conflict affecting a shared characteristic would be the reproductive organs in males and females. This study investigates how artificial selection on the female-specific trait egg size influences male reproductive success, particularly male testes morphology, by analysing the density and number of seminiferous tubules within the testes. It was found that selection on female reproductive investment has a concordant effect on male reproductive success, by demonstrating increased density of seminiferous tissue in the testes of males originating from high investment selection lines. Interestingly, a non-significant trend suggested that the density of seminiferous tissue influences testes size in a negative way. This study therefore provides evidence that female-specific selection on reproductive investment influences testes morphology in males, and that testes size depends on more than sperm-producing tissue.
Introduction
In a perfect model population, sexual selection would be expected to always act in favour of each sex and promote males and females to reach their sex-specific optima, resulting in highly dimorphic sexes.
In reality however, sexual selection can act strongly on males and females, thereby driving the sex- specific optima for a certain trait further apart, causing the genetic interests of both sexes to diverge, also known as sexual conflict (Chapman et al. 2003; Berger et al. 2016). Sexual conflict can be classified into two main categories, interlocus and intralocus conflict (Chapman et al. 2003). Interlocus conflicts are caused by a conflict between sexes over the result of male-female interactions, with each sex having a different optimal outcome (Rice and Holland 1997), for example mating frequency or parental effort (Chapman et al. 2003) where it is best for males to mate with many different females and exert no paternal care, whereas the opposite would be best for the females. In contrast to this, intralocus
2 | P a g e conflict is caused by antagonistic sexual selection on a trait that is genetically correlated between the sexes, leading to different fitness optima in males and females, thus disabling each sex to reach its adaptive evolutionary optimum (Lande 1980; Parker and Partridge 1998; Chapman et al. 2003; Poissant et al. 2010). If selection acts on a sex-limited gene, however, it can promote sex-independent evolution in males and females, with both sexes diverging towards their individual optimum (Rhen 2000).
Ultimately, intralocus conflict is expected to force both sexes to diverge further and further until they eventually develop a separate genetic basis to reach their sex-specific optima expressed in sexual dimorphism (Poissant et al. 2010). Following this assumption, sexual dimorphism is often regarded as an indicator for past intralocus sexual conflict in a species (Cox and Calsbeek 2009).
Male and female reproductive functions are prime examples for sexual dimorphism, as they are typically expressed in a sex-limited way (Pick et al. 2017). If sex-limited selection acts on reproductive functions with independent genetic bases for both sexes, it will promote the function in one sex while having little effect on the fitness of the other sex (Pick et al. 2017). However, if the genetic basis is shared between males and females, sex-limited selection may have a considerable impact on the other sex (Fischer et al. 2009; Pick et al. 2017). One prominent example for sexual dimorphism is the divergent morphology of gonads in males and females. Females in many bird species show an asymmetric development of gonads with only the left ovary developing fully and the right ovary regressing after the initial developmental stage (Stanley and Witschi 1940; Intarapat and Stern 2013;
Calhim and Montgomerie 2015). Males on the other hand, develop testes on both sides, but some species show asymmetries in testes size with the left testis generally being enlarged (Stanley and Witschi 1940; Møller 1994). Relative testes size has often been used as a proxy for male reproductive success and fertility in previous studies (Preston et al. 2003; Ramm and Schärer 2014) with larger testes usually being assumed to produce more sperm (Møller 1988, 1989, 1994; Calhim and Birkhead 2009;
Ramm and Schärer 2014). However, recent findings suggest that testes size alone is not the only variable influencing reproductive success, and propose testes morphology to have a significant impact on fertility in males (Ramm and Schärer 2014; Firman et al. 2015; Pick et al. 2017).
3 | P a g e To investigate the effect selection on female reproductive investment has on male reproductive success, Pick et al. (2017) have established two divergent, replicate artificial selection lines for female egg investment in the Japanese quail (Coturnix japonica). The first line selected for females with high reproductive investment, whereas the second line selected for females with low reproductive investment (Pick et al. 2016b). Female egg investment gives a good representation of female resource investment in their offspring and therefore a representative indication of offspring fitness (Krist 2011;
Pick et al. 2016a,c) and in this case male reproductive success (Pick et al. 2017). They found that female-limited selection on egg investment positively affected not only female gonad size on the left side (Pick et al. 2016b), but also male reproductive success and that males that originated from high maternal selection lines showed a higher testes asymmetry with the left testis being enlarged, which suggests a positive genetic correlation between the gonadal asymmetry in both sexes (Pick et al. 2017).
This study builds on the findings of Pick et al. (2017) with the aim to investigate the role artificial selection on female reproductive investment has on male fertility by not only considering the testes size but the morphology and functionality of the testes. Besides the testes size, the density of sperm- producing tissue within the testes has been proposed to play an important role in determining male fertility (Ramm and Schärer 2014). The testes serve a diverse range of functions, including, but not limited to, sperm and hormone production (Firman et al. 2015). The sperm-producing part of the testes contains mainly seminiferous tubules that are composed of seminiferous tissue with a basal lamina and lumina in the middle (see Fig. 1). The seminiferous tubules contain germ cells that develop into spermatozoa before they are passed through the lumen to the efferent ducts to later mature into single sperm (Wistuba et al. 2007; Firman et al. 2015).
4 | P a g e As males from high maternal investment selection lines have been found to show higher reproductive success than males from low maternal investment selection lines (Pick et al. 2017), and testes size and the density of sperm-producing tissue are positively associated with male fertility and reproductive success (Preston et al. 2003; Ramm and Schärer 2014; Firman et al. 2015), it is therefore hypothesised that males from high maternal investment lines have a higher amount of sperm-producing tissue than males from low maternal investment lines and that the amount of sperm-producing tissue positively affects male fertility, measured by the proxies of sperm concentration and testes mass. It is predicted that 1) males from high maternal investment lines have a higher percentage of seminiferous tissue than males from low maternal investment lines, that 2) males from high maternal investment lines have a higher number of seminiferous tubules than males from low maternal investment lines, that 3) males with a higher percentage of seminiferous tissue and/or number of seminiferous tubules have a higher sperm concentration, and that 4) males with a higher percentage of seminiferous tissue and/or number of seminiferous tubules have a higher testes mass.
Figure 1: Labelled description of a single seminiferous tubule of a testis slide under a 2x magnification, showing the seminiferous tissue, lumen, basal lamina and intermediate area.
5 | P a g e For the purpose of simplification, males that originated from high reproductive investment lines will be referred to as “high males” and males that originated from low reproductive investment lines will be referred to as “low males” in the following.
Material and methods
A base study population of captive Japanese quail has been used to establish artificial selection lines for high and low maternal investment (Pick et al. 2016c). The selection criterion for maternal investment was relative egg size, obtained by measuring absolute egg size and female body size. For each of the four generations and replicates, the most extreme 50% regarding high and low maternal egg investment were selected for further studies and the offspring from these eggs were isolated and artificially incubated simultaneously to minimise the influence of environmental factors. Males and females from each breeding pair were not related to each other in the sense of not sharing the same parents or grandparents. For more detailed information on the artificial selection procedure see Pick et al. (2016c).
During previous work, 20 males (10 males from high reproductive investment lines and 10 males from low reproductive investment lines) have been euthanised and histological sections of different levels of the testes have been produced (see Pick et al. 2017). All procedures were conducted under licenses provided by the Veterinary Office of the Canton of Zurich, Switzerland (permit numbers 195/2010;
14/2014; 156). The complete sample size consisted of histological slides of 20 individuals with a 2x magnification, 10 slides per male with 5 slides per testis.
The slides were analysed using the three different methods outlined below using the programme ImageJ, to determine the density and quantity of seminiferous tubules in the testes. The analysis has been conducted blindly in regard to the males’ treatment group and any other additional information to avoid observer bias. After the initial analysis was complete, additional information has been linked to the results in order to progress with the statistical analyses. Any additional data used in the
6 | P a g e statistical analyses, that were not obtained by the methods outlined below, have been collected during previous studies (i.e. sperm concentration and testes mass). Testes mass in grams has been obtained by dissecting the euthanised males, extracting the testes and weighing them individually (Pick et al.
2016b, 2017). Sperm concentration has been obtained by counting the number of sperm per millilitre of semen (Bladen, 2019 unpublished data).
Quantification of the seminiferous tubule number using the counting method
This method was used to count the number of seminiferous tubules per slide. The total sample size was 200 slides, consisting of 20 males and 10 slides per male with 5 slides per testis. Each seminiferous tubule was counted once, even if it had several lumina because this can be attributed to the layering of tubules and the position of the cut. Tubules were counted as individuals when they showed a distinct outer layer, even if they touched other tubules. Seminiferous tubules at the edge of the slide were counted if they had a visible lumen; if the lumen was not visible, even though the visible tissue was relatively large, it was not counted as a tubule.
Quantification of the seminiferous tubule area using the circling method
This method was used to circle all seminiferous tubules per slide using the programme ImageJ (Rasband 2018), including tissue without visible lumen on the edges of the slides. The total sample size for this method was 40 slides, consisting of 20 males and 2 slides per male with one slide per testis.
This method was very time intensive and has therefore only been applied to slide number three of each testis, because this slide represented the middle section of the testis. Using the middle slide of each series provided a comparability between testes because the same layer for each testis has been analysed. This method included the lumen of seminiferous tubules as well as smaller tissue pieces at the edges of the slide but excluded any smaller cut-offs in the intermediate area between tubules because these were too small to circle. This method provided few but highly accurate measurements.
To obtain the percentage of seminiferous tubule area (including the lumina) per slide, the following formula was applied:
7 | P a g e Total tubule area = Seminiferous tubule area / Total image area
More detailed information on the analysis process in ImageJ can be found in the Appendix section.
Figure 2 shows an example of a finished slide with the circling method applied to all seminiferous tubules on the slide.
Quantification of the seminiferous tubule area using the adjusted threshold method
This method used an adjusted threshold to measure the seminiferous tubule percentage per slide using the programme ImageJ. The total sample size for this method was 200 slides, consisting of 20 males and 10 slides per male with 5 slides per testis. The use of an adjusted threshold for each individual slide, instead of a set threshold for all slides, was necessary because of the difference in exposure and colour intensity of the slides. Using an adjusted threshold was justifiable because the analysis has been conducted blindly regarding the males’ treatment group and has been completed twice for the whole sample size with one week in between measurements.
Figure 2: Testis slide with all seminiferous tubules circled in yellow using the circling method.
8 | P a g e To progress with the statistical analyses, the arithmetic mean of both measurements has been used.
This method included smaller cut-offs in the intermediate area between seminiferous tubules and excluded the lumina, as only the coloured tissue was measured. The slides showed high variation in exposure, brightness, colour intensity and artefacts – some were very distinct with the threshold setting, whereas others showed a fisheye effect and highly varied in optimal threshold setting depending on the focus. For any slides that showed such effects, the threshold was adjusted as accurately as possible to capture as much of the sperm-producing tissue as possible.
More information on the analysis process in ImageJ can be found in the Appendix section. Figure 3 shows an example of a testis slide with black seminiferous tissue pixels and white non-tissue pixels. To obtain the percentage of seminiferous tubule area (excluding the lumina) per slide, the following formula was applied:
Total tissue percentage = Tissue pixels / (Tissue pixels + Non-tissue pixels)
Figure 3: Testis slide with seminiferous tissue coloured black and non-seminiferous tissue and lumina coloured white, using the adjusted threshold method to capture the amount of seminiferous tissue as accurately as possible.
9 | P a g e Statistical analyses
The data was normally distributed, therefore parametric tests have been used for the following analyses using the software R (Version 3.5.2; R Core Team, 2018). The repeatability of the threshold measurements has been tested to determine the quality of measurements by comparing the variation within an individual to the overall variation of the measurements, thereby determining the accuracy of the measurements and how much variation can be explained by observer error, using the rptR package (Stoffel et al. 2017). Additionally, the correlation between the arithmetic mean of the threshold measurements and the circling measurements have been tested for by conducting a Pearson’s correlation test. For the correlation test between threshold and circling measurements, the
arithmetic mean of the third slide of each series has been used as the circling method has only been conducted on slide number three of each testis. The repeatability test for the adjustable threshold analysis showed that the two measurements have a very high repeatability, using the ‘Gaussian’ setting and bootstrapping it 1000 times (R=0.948, SD=0.023, p=8.44e-53). The arithmetic mean of the seminiferous tubule percentage obtained by the adjusted threshold method showed only low correlation with the seminiferous tubule percentage obtained by the circling method (r=0.220, df=38, t=1.392, p=0.172).
To test for the four predictions, firstly, if high males have a higher percentage of seminiferous tubules than low males, linear models have been used with ‘tissue percentage’ as response variable and
‘maternal investment’ as effect variable (N=20 males). The model has been run with the seminiferous
tissue percentage obtained by the circling method and the adjusted threshold method. Additionally, the model has been run for the two testes sides using the seminiferous tissue percentage obtained by the circling method for the left and right testis as response variable and ‘maternal investment’ as the effect variable (N=20 males).
10 | P a g e Secondly, to test if high males have a higher number of seminiferous tubules than low males, a linear model has been used with ‘number of tubules’ as response variable and ‘maternal investment’ as effect variable (N=20 males).
Thirdly, to investigate if males with a higher seminiferous tissue percentage or a higher number of seminiferous tubules have a higher sperm concentration, linear mixed models have been used with
‘sperm concentration’ as response variable, ‘tissue percentage’ or ‘number of tubules’ as effect variable and ‘maternal investment’ as fixed effect (N=20 males). The model has been run with the seminiferous tissue percentage obtained by the circling and the adjusted threshold method. Maternal investment has been set as a fixed effect as this factor is expected to influence sperm concentration.
Lastly, to test if males with a higher seminiferous tissue percentage or a higher number of seminiferous tubules have a higher testes mass, linear mixed models have been used with ‘testes mass’ as response variable, ‘tissue percentage’ or ‘number of tubules’ as effect variable and ‘maternal investment’ as
fixed effect (N=20 males). The model has been run with the seminiferous tissue percentage obtained by the circling and the adjusted threshold method. Maternal investment has been set as a fixed effect as this factor is expected to influence sperm concentration.
Statistical significance has been determined using F statistics with df = 18 for linear models and df = 17 for linear mixed models.
11 | P a g e
Results
High males showed a significantly higher amount of seminiferous tissue compared to low males, when using the tissue percentage obtained by the circling method (p=0.029, df=18, N=20; Fig. 4a). When analysing the separate testes sides, there was a trend for the left testes of high males showing a higher tissue percentage than low males but with no statistical significance (Fig. 4b), whereas the analysis for the right testes showed that high males have a significantly higher amount of seminiferous tissue than low males (p=0.049, df=18, N=20; Fig. 4c).
Figure 4a: The influence maternal investment has on the total seminiferous tissue percentage. Seminiferous tissue percentage is measured from 0% - 200% as it displays the combined seminiferous tissue percentage of left and right testes. Maternal investment is categorised as ‘high’ and ‘low’.
Figure 4b: The influence maternal investment has on the total seminiferous tissue percentage in the left testes. Seminiferous tissue percentage is measured from 0% - 100%. Maternal investment is categorised as
‘high’ and ‘low’.
12 | P a g e In concordance with the results described above, high males showed a relatively higher amount of seminiferous tissue compared to low males when using the tissue percentage obtained by the adjusted threshold method but did not reach statistical significance.
When comparing the number of seminiferous tubules per testis between high and low males, against the prediction, low males showed a relatively higher number of tubules than high males, but with no statistical significance reached.
Contrary to the prediction that males with a higher percentage of seminiferous tubules have a higher sperm concentration, the results of the linear mixed models showed no significant effect for neither tissue percentage obtained by the circling method, nor for the adjusted threshold method. The same can be said about the influence of the number of seminiferous tubules on the sperm concentration.
Tissue percentage and number of tubules both showed a positive trend in correlation to sperm concentration but did not reach statistical significance.
The seminiferous tissue percentage obtained by the circling method showed a positive impact on the testes mass, although with no statistical significance (Fig. 5). In contrast to the prediction and the results from the model using the circling method, the seminiferous tissue percentage obtained by the adjusted threshold method showed a negative, yet not significant, relationship with the testes mass Figure 4c: The influence maternal investment has on the total seminiferous tissue percentage in the right testes.
Seminiferous tissue percentage is measured from 0% - 100%. Maternal investment is categorised as ‘high’ and ‘low’.
13 | P a g e (Fig. 6). In line with this, the number of seminiferous tubules demonstrated a negative relationship with the testes mass without statistical significance (Fig. 7).
Figure 5: The influence the total seminiferous tissue percentage obtained by the circling method has on the total testes mass with maternal investment incorporated as fixed effect. Seminiferous tissue percentage is measured from 0% - 200% as it displays the combined seminiferous tissue percentage of left and right testes.
Figure 6: The influence the total seminiferous tissue percentage obtained by the adjusted threshold method has on the total testes mass with maternal investment incorporated as fixed effect. Seminiferous tissue percentage is measured from 0% - 200% as it displays the combined seminiferous tissue percentage of left and right testes.
14 | P a g e
Discussion
It has been shown that the artificial selection for high and low egg investment in females positively influences the size of the left ovary in females from high investment selection lines (Pick et al. 2016b).
Simultaneously, males from high maternal investment lines showed a higher testes asymmetry with the left testis enlarged, which was positively linked to increased reproductive success (Pick et al. 2017).
It has been proposed that the two testes might differ in function, with the left testis serving as the main functional testis (Møller 1994; Calhim and Birkhead 2009; Pick et al. 2017). Recent studies showed that gene expression in the left gonad is higher than in the right for both sexes in the early developmental stage (Intarapat and Stern 2013, 2014), which could potentially lead to differences in functionality in the adult stage in male testes (Pick et al. 2017).
This study found that high males show a significantly higher percentage of seminiferous tissue compared to low males, which coupled with previous findings (Pick et al. 2017), indicates a positive relationship between the quantity of seminiferous tissue and the reproductive success of males. The
Figure 7: The influence the number of seminiferous tubules has on the total testes mass with maternal investment incorporated as fixed effect.
15 | P a g e more in-depth analyses of the separate testes sides confirmed that high males show a higher seminiferous tissue percentage on both sides, with a significant difference between high and low males in the right testis. High males did not show a significantly higher seminiferous tissue percentage in the left testis, however, this could be attributed to the small sample size as only a limited amount of slides has been included in the analysis. These results do not suggest that the left testis serves a different function than the right testis or that is has a higher level of functionality in terms of sperm-production.
Nevertheless, the results comply with previous findings that sperm competition selects for increased density of seminiferous tissue in more competitive males (Pitnick et al. 2001; Lüpold et al. 2009, 2011;
Firman et al. 2015), which in this case applies to high males having higher reproductive success than low males.
Besides testes size and composition, sperm concentration and sperm size is often used as a proxy for male fertility (Parker 1982; Ramm and Schärer 2014). When looking at the influence, density of seminiferous tissue and number of seminiferous tubules have on the sperm concentration in males, the results showed a positive trend of sperm concentration increasing with increasing seminiferous tissue amount. Although the results did not reach a statistically significant level, they are in line with the assumption that sperm concentration is positively correlated to the amount of sperm-producing tissue (Ramm and Schärer 2014; Firman et al. 2015), thus supporting the previous findings that males from high maternal investment lines have a higher fertility compared to males from low maternal investment lines (Pick et al. 2017).
Contrary to the prediction that high males have a higher number of seminiferous tubules, especially in the left testis, the results showed a trend that low males have in fact a relatively higher number of seminiferous tubules. Nevertheless, it might not be the sole number of seminiferous tubules that is important for determining fertility, but other attributes like seminiferous tubule diameter and the lumina size within the tubules, as these have been shown to vary under different conditions (Artoni et al. 1997, 1999). Low males may have a higher number of seminiferous tubules, but high males could potentially have longer tubules, a bigger tubule diameter and smaller lumina, thus presenting a
16 | P a g e strategic development tailored to the quality of the male (Firman et al. 2018). It has been shown that sperm competition influences the size and shape of sperm significantly, with longer sperm requiring larger and wider seminiferous tubules (Pitnick 1996; Lüpold et al. 2009). However, as none of these variables were included in the analyses, it is, at this point, purely speculative and opens up the prospect of follow-up questions that could be investigated further.
As previous studies found, high males displayed a testes asymmetry with the left testis being larger than the right, which was not found in low males (Pick et al. 2017). Therefore, it would be expected that the left testes of high males contain more seminiferous tissue and a higher number of seminiferous tubules indicating an additional or different function of those testes (Calhim and Birkhead 2009; Pick et al. 2017). The negative correlation between the percentage of sperm-producing tissue and the testes mass, and the negative correlation between the quantity of seminiferous tubules and testes mass, contradict the prediction that an increase in sperm-producing tissue leads to an increase in testes size (Gage 1994; Firman and Simmons 2008; Firman et al. 2015). Although the tissue percentage obtained by the circling method showed a positive influence on the testes mass (Fig. 5), the tissue percentage obtained by the threshold method showed the opposite (Fig. 6). In this case, the threshold method might be the more suitable choice as it displays the percentage of only seminiferous tissue, whereas the circling method displays the percentage of seminiferous tissue and lumina. When looking at the comparability between the two sets of adjusted threshold measurements, the high repeatability suggests that the observer accuracy was high and that the variation within an individual was low compared to the variation between individuals. It also means that only a very small proportion of the variation can be explained by observer inaccuracy and that most of the variation originated from an unknown variable. Although a high correlation between the adjusted threshold method measurements and the circling method measurements was expected, the resulting low correlation could be accounted for by the differences between the two methods. The circling method included all lumina, whereas the threshold method excluded them. A variation in lumina size could be an
17 | P a g e influencing factor, as well as a possible inaccuracy of the threshold measurements when the threshold had to be adjusted to a suboptimal level to compensate for lighter or darker areas of a slide.
As testes mass is often used as a proxy for male fertility and reproductive success in competition with other males (Ramm and Schärer 2014; Firman et al. 2015), it would be expected that the density of seminiferous tissue is positively correlated to the testes size. However, the results of this study indicate that enlarged testes in high males do not originate from a higher accumulation of sperm-producing tissue and number of seminiferous tubules in the testes. Recent studies suggest that sperm competition selects for a higher density of seminiferous tissue to increase testicular function, rather than selecting for increased testes size (Pitnick et al. 2001; Firman et al. 2015), which could mean that testes size itself is not necessarily correlated to the amount of sperm-producing tissue. Interestingly, Pick et al. (2017) found no significant relationship between total testes mass and fertilisation success in males despite a strong difference in male reproductive success between high and low males, but instead found a significant difference in testes asymmetry.
This raises the question why the left testes of high males are enlarged and what the underlying cause of the testes asymmetry is, as the density of sperm-producing tissue seemed to have no promoting effect on testes mass. It could be possible that testes mass is dependent on different factors, which not necessarily serve the function of increasing fertility (Ramm and Schärer 2014), like lumina size or the density of the interstitial tissue that contains blood vessels and serves endocrine functions (Wistuba et al. 2007). This would suggest that testes mass itself might not be the best indicator for male fertility and that other factors, like testes morphology and structure are more influential than size when investigating what target sexual selection acts on to increase reproductive success (Pitnick 1996; Ramm and Schärer 2014). Møller (1994) proposed another explanation for testes asymmetry, namely that testes size composes a handicap for males and that only males of high quality can afford to expend the cost of having one enlarged testis (see Zahavi 1975 for handicap theory). If it was assumed that testes asymmetry is costly, males in poor condition may be unable to afford an enlarged left testis and have a smaller testes asymmetry with the left and right testes being of similar, smaller
18 | P a g e size (Møller 1994). For good quality males, the smaller right testis would act as a back-up that grows in size if the larger left testis with higher functionality failed (Lake 1984). This suggests that the right testis purely serves the purpose of compensating for a loss in functionality in the left testis. However, this theory has been contradicted by a study on zebra finches, that found that asymmetry in testes size did not reflect the quality of males (Birkhead et al. 1998). If there is a developmental advantage to increasing the size of only one testis, it could potentially be more energetically advantageous to increase the total testes mass by increasing only one testis, or physiologically beneficial by increasing only one testis to increase testes mass in a limited space (Calhim and Montgomerie 2015), or as an adaptation for flight (Witschi 1935). Calhim and Birkhead (2009) proposed that testes asymmetry is the result of a trade-off between increased testicular function and constrained space within the body cavity. They showed that testis compensation (Møller 1994) can occur naturally in birds and becomes effective if one testis fails (Calhim and Birkhead 2009). The most recent and consistent suggestion for testes asymmetry is that testes asymmetry in males is a by-product of sexual selection acting on the female reproductive organs, causing female ovaries to develop only on the left side and regress on the right (Stanley and Witschi 1940; Calhim and Montgomerie 2015). The results from this and previous studies (Pick et al. 2017) support this theory, as Pick et al. (2017) found that artificial selection on female reproductive investment acts genetically concordant on male reproductive success, with the selection having a positive effect on the size of the left ovary in females and testes asymmetry with the left testis enlarged in males of high investment lines. Another suggestion is that the artificial selection for egg size did not only select for females to produce larger eggs, but to invest more resources into reproduction (Fischer et al. 2006, 2009), thus it could be assumed that due to the genetic concordance between males and females, high males present with an enlarged left testis because they invest more resources into reproduction compared to low males, which is corresponding with the findings of this study that high males have a significantly higher percentage of sperm- producing tissue.
19 | P a g e In conclusion, this study demonstrated that sex-specific selection on the reproductive trait egg size in females, positively affects the density of seminiferous tissue in male testes with a positive correlation to sperm concentration. Selection for increased reproductive investment in females therefore correlates with increased reproductive success in males, expressing itself in increased density of seminiferous tissue rather than total testes mass. Interestingly, the amount of seminiferous tissue seemed to impact testes mass in a negative way, thus suggesting that the increased relative testes mass in males from high reproductive investment lines originates from something other than sperm- producing tissue. This is in compliance with the recent suggestions that testes mass alone might not be a sufficient indicator for male fertility and that testes composition and structure play a vital role in assessing how sexual selection acts on male reproductive success.
20 | P a g e
Appendix
Quantification of the seminiferous tubule area using the circling method
To analyse a slide in ImageJ, the image needs to be loaded into the programme. The first step is calibrating the scale by taking repeated measurements of the graticule using the ‘straight line’ tool and taking the mean of these measurements. The mean needs to be entered in pixels, as well as the actual length of the graticule to obtain a pixels/µm ratio with a pixel aspect ratio of 1 (here: 172 pixels = 500 µm = 0.344 pixels/µm). The setting can be set to ‘global’ if all slides have the same scale. The next step is setting the measurements for the analysis, select ‘area’ to obtain areal measurements in the later process. Using the ‘freehand selection’ tool, each tubule needs to be circled as accurately as possible, following the basal lamina (outer layer of the tubule) until the starting point is reached again. After each completed tubule circling, the measurement needs to be added to the ROI manager pressing ‘t’.
Progress can be tracked by selecting the option ‘show all’ in the ROI manager. Once all tubules have been circled, all measurements in the ROI manager need to be selected and using the option ‘measure’
the area of each tubule will be displayed and can be exported and summed up in Microsoft Excel to generate the seminiferous tubule area.
Quantification of the seminiferous tubule area using the adjusted threshold method
To analyse a slide in ImageJ, the image needs to be loaded into the programme and set to greyscale by changing the image type to ‘8 bit’. To adjust the threshold individually for each slide, the lower sliding
scale in the threshold setting needs to be adjusted until all pixels that should contain seminiferous tissue appear in a red colour. After applying the threshold setting to the image, all pixels containing tissue appear black and all non-tissue containing pixels appear white. The number of pixels for seminiferous tissue can be found using the histogram tool that contains a list of the number of pixels for all 256 shades of grey. The output displays binary values for 0 (all white non-tissue pixels) and 255 (all black tissue pixels).
21 | P a g e Acknowledgements
I thank Dr. Barbara Tschirren and Dr. Richard Svanbäck for the opportunity to conduct my research under their guidance. I thank Matt Ryan for the ongoing support throughout the project.
References
Artoni, S. M. B. B., A. M. Orsi, T. L. L. Carvalho, and R. A. Lopes. 1997. The Annual Testicular Cycle of the Domestic Quail (Coturnix coturnix japonica). Anat. Histol. Embryol. 26:337–339.
Artoni, S. M. B. B., A. M. Orsi, T. L. L. Carvalho, C. A. Vicentini, and M. A. Stefanini. 1999. Seasonal Morphology of the Domestic Quail (Coturnix coturnix japonica) Testis. Anat. Histol. Embryol.
28:217–220.
Berger, D., I. Martinossi-Allibert, K. Grieshop, M. I. Lind, A. A. Maklakov, and G. Arnqvist. 2016.
Intralocus sexual conflict and the tragedy of the commons in seed beetles. Am. Nat. 188:E98– E112.
Birkhead, T. R., F. Fletcher, and E. J. Pellatt. 1998. Testes asymmetry, condition and sexual selection in birds: An experimental test. Proc. R. Soc. B Biol. Sci. 265:1185–1189.
Bladen, C. 2019. Unpublished data.
Calhim, S., and T. R. Birkhead. 2009. Intraspecific variation in testis asymmetry in birds: Evidence for naturally occurring compensation. Proc. R. Soc. B Biol. Sci. 276:2279–2284.
Calhim, S., and R. Montgomerie. 2015. Testis asymmetry in birds: The influences of sexual and natural selection. J. Avian Biol. 46:175–185.
Chapman, T., G. Arnqvist, J. Bangham, and L. Rowe. 2003. Sexual conflict. Trends Ecol. Evol. 18:41–47.
Cox, R. M., and R. Calsbeek. 2009. Sexually antagonistic selection, sexual dimorphism, and the resolution of intralocus sexual conflict. Am. Nat. 173:176–187.
Firman, R. C., F. Garcia-Gonzalez, L. W. Simmons, and G. I. André. 2018. A competitive environment influences sperm production, but not testes tissue composition, in house mice. J. Evol. Biol.
31:1647–1654.
Firman, R. C., F. Garcia-Gonzalez, E. Thyer, S. Wheeler, Z. Yamin, M. Yuan, and L. W. Simmons. 2015.
Evolutionary change in testes tissue composition among experimental populations of house mice.
Evolution (N. Y). 69:848–855.
Firman, R. C., and L. W. Simmons. 2008. The frequency of multiple paternity predicts variation in testes size among island populations of house mice. J. Evol. Biol. 21:1524–1533.
Fischer, K., A. N. M. Bot, P. M. Brakefield, and B. J. Zwaan. 2006. Do mothers producing large offspring have to sacrifice fecundity? J. Evol. Biol. 19:380–391.
22 | P a g e Fischer, K., K. Zimmer, and N. Wedell. 2009. Correlated responses to selection on female egg size in
male reproductive traits in a butterfly. Evol. Ecol. 23:389–402.
Gage, M. J. G. 1994. Association between body size, mating pattern, testis size and sperm lengths across butterflies. Proc. R. Soc. B Biol. Sci. 258:247–254.
Intarapat, S., and C. D. Stern. 2014. Left-right asymmetry in chicken embryonic gonads. J. Poult. Sci.
51:352–358.
Intarapat, S., and C. D. Stern. 2013. Sexually Dimorphic and Sex-Independent Left-Right Asymmetries in Chicken Embryonic Gonads. PLoS One 8:1–8.
Krist, M. 2011. Egg size and offspring quality: a meta-analysis in birds. Biol. Rev. 86:692–716.
Lake, P. E. 1984. Male genital organs. Pp. 1–61 in A. S. King and J. McLelland, eds. Form and function in birds. Vol. 1. Academic Press, London.
Lande, R. 1980. Sexual Dimorphism, Sexual Selection, and Adaptation in Polygenic Characters.
Evolution (N. Y). 34:292–305.
Lüpold, S., G. M. Linz, J. W. Rivers, D. F. Westneat, and T. R. Birkhead. 2009. Sperm competition selects beyond relative testes size in birds. Evolution (N. Y). 63:391–402.
Lüpold, S., J. Wistuba, O. S. Damm, J. W. Rivers, and T. R. Birkhead. 2011. Sperm competition leads to functional adaptations in avian testes to maximize sperm quantity and quality. Reproduction 141:595–605.
Møller, A. P. 1994. Directional selection on directional asymmetry: Testes size and secondary sexual characters in birds. Proc. R. Soc. B Biol. Sci. 258:147–151.
Møller, A. P. 1989. Ejaculate Quality, Testes Size and Sperm Production in Mammals. Funct. Ecol. 3:91– 96.
Møller, A. P. 1988. Testes size, ejaculate quality and sperm competition in birds. Biol. J. Linn. Soc.
33:273–283.
Parker, G. A. 1982. Why are there so many tiny sperm? Sperm competition and the maintenance of two sexes. J. Theor. Biol. 96:281–294.
Parker, G. A., and L. Partridge. 1998. Sexual conflict and speciation. Philos. Trans. R. Soc. London Biol.
Sci. 353:261–274.
Pick, J. L., C. Ebneter, P. Hutter, and B. Tschirren. 2016a. Disentangling genetic and prenatal maternal effects on offspring size and survival. Am. Nat. 188:628–639.
Pick, J. L., P. Hutter, C. Ebneter, A. K. Ziegler, M. Giordano, and B. Tschirren. 2016b. Artificial selection reveals the energetic expense of producing larger eggs. Front. Zool. 13:1–10. Frontiers in Zoology.
Pick, J. L., P. Hutter, and B. Tschirren. 2017. Divergent artificial selection for female reproductive investment has a sexually concordant effect on male reproductive success. Evol. Lett. 1:222–228.
Pick, J. L., P. Hutter, and B. Tschirren. 2016c. In search of genetic constraints limiting the evolution of egg size: Direct and correlated responses to artificial selection on a prenatal maternal effector.
Heredity (Edinb). 116:542–549. Nature Publishing Group.
23 | P a g e Pitnick, S. 1996. Investment in testes and the cost of making long sperm in Drosophila. Am. Nat.
148:57–80.
Pitnick, S., G. T. Miller, J. Reagan, and B. Holland. 2001. Males’ evolutionary responses to experimental removal of sexual selection. Proc. R. Soc. B Biol. Sci. 268:1071–1080.
Poissant, J., A. J. Wilson, and D. W. Coltman. 2010. Sex-specific genetic variance and the evolution of sexual dimorphism: A systematic review of cross-sex genetic correlations. Evolution (N. Y). 64:97– 107.
Preston, B. T., I. R. Stevenson, J. M. Pemberton, D. W. Coltman, and K. Wilson. 2003. Overt and covert competition in a promiscuous mammal: The importance of weaponry and testes size to male reproductive success. Proc. R. Soc. B Biol. Sci. 270:633–640.
R Core Team. 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria.
Ramm, S. A., and L. Schärer. 2014. The evolutionary ecology of testicular function: Size isn’t everything.
Biol. Rev. 89:874–888.
Rasband, W. J. 2018. ImageJ. U.S. National Institutes of Health, Bethesda, Maryland, USA.
Rhen, T. 2000. Sex-limited mutations and the evolution of sexual dimorphism. Evolution (N. Y). 54:37– 43.
Rice, W. R., and B. Holland. 1997. The enemies within: intergenomic conflict, interlocus contest evolution (ICE), and the intraspecific Red Queen. Behav. Ecol. Sociobiol. 41:1–10.
Stanley, A. J., and E. Witschi. 1940. Germ cell migration in relation to asymmetry in the sex glands of hawks. Anat. Rec. 76:329–342.
Stoffel, M. A., S. Nakagawa, and H. Schielzeth. 2017. rptR: Repeatability estimation and variance decomposition by generalized linear mixed-effects models. Methods Ecol Evol 8:1639–1644.
Wistuba, J., J.-B. Stukenborg, and C. M. Luetjens. 2007. Mammalian Spermatogenesis. Funct. Dev.
Embryol. 1:99–117.
Witschi, E. 1935. Origin of asymmetry in the reproductive system of birds. Am. J. Anat. 56:119–141.
Zahavi, A. 1975. Mate Selection-A Selection for a Handicap. J. Theor. Biol. 53:205–214.
