Anyone who has worked with anoles in the field can tell you that during the breeding season, you can see these lizards mating pretty frequently. In fact, brown anoles are so promiscuous they have one of the highest rates of multiple paternity in vertebrates! In addition to this, females can store sperm for months at a time, which suggests that this group probably experiences a lot of sperm competition. We know from other groups that sperm competition can lead to the evolution of lots of interesting and diverse traits, but we know relatively little about how sperm competition targets these traits in anoles (though see Klaczko et al’s 2015 paper about the rate of genital evolution in anoles).
Some of our previous work has shown that sperm size and sperm count are correlated with reproductive success in the lab, but it’s unclear how these important components of male reproductive success evolve. To explore this, we collected data on testis size (as a proxy for sperm production/sperm count), and sperm morphology (measuring the sperm head, midpiece and tail) for 26 species of anoles. We hypothesized that if testis size had a higher rate of evolution than sperm morphology, sperm competition likely targets sperm count over sperm size.
Rates of evolution of testis size, body size and sperm morphology next to an anole sperm.
In fact, we found that testis size evolves much faster than sperm morphology, confirming our hypothesis. We also found that within the sperm, the midpiece (which contains the mitochondria) evolves 2-3 times faster than the rest of the sperm cell. This is maybe unsurprising, considering that the midpiece size is correlated with reproductive success in brown anoles, and is often associated with sperm longevity and sperm velocity in other species.
You can read the pre-print from JEB below!
Kahrl, AF, Johnson, MA, and Cox, RM. 2019.Rapid evolution of testis size relative to sperm morphology suggests that postcopulatory selection targets sperm number in Anolis lizards. Journal of Evolutionary Biology.
Theory predicts that males should invest more in ejaculate production when the likelihood of sperm competition is high, thereby increasing the chance of fertilization. However, ejaculates can be energetically costly, and increased investment into sperm production should only occur if there are fitness benefits associated with that increased investment. Growing experimental evidence suggests that sperm traits respond plastically to social environment. However, it is not known whether fine-scale spatial variation in the local density of male competitors or potential female mates corresponds to individual variation in ejaculate production.
Island population with capture records of males (blue) and female (red) anoles.
Matt Kustra of the Cox lab examined a wild population to test the prediction that, as the risk of sperm competition increases (i.e., higher local density of male competitors), males will increase their total investment in their ejaculates (sperm count). He also tested for correlations between sperm morphology, specifically midpiece size and local density.
To do this, he and the Cox lab collected wild adults from an island population in Florida. They generated a map of each tree on the island using ArcGIS, then marked the location of males and females on this map. Using the kernel density function, they estimated the local density of individual males by taking into account all conspecific adults that were captured within a 5.8 m radius of an individual’s own capture location.
Matt found that length of the sperm midpiece increased with local density, whereas length of the sperm head and sperm count decreased with local density. Contrary to his predictions, he found that total investment in sperm count decreased with local density. This could be because males in high density environments have depleted their sperm stores because they have more opportunities to mate, or it could be because males are investing less per ejaculate if mating frequency is higher.
These findings indicate that fine-scale differences in local density within a wild population can affect sperm count and various sperm phenotypes. In the future, the Cox lab hopes to measure fitness in this populations to understand how sperm phenotypes shape individual reproductive success.
Anolis RPM muscle cross-section. The darkened area in the middle of the muscle shows white blood cell infiltration, which indicates damaged tissue.
Many anole biologists have spent a lot of time, money, sweat, and tears collecting behavioral data in the field. These estimates of behavior are very important for understanding how sexual selection operates and how the structures associated with those behaviors evolve, but they are notoriously hard to collect. Kyle Martin and his collaborators decided to try to determine if there might be a better and more accurate proxy for the frequency of specific behavioral displays. As structures are used, so are the muscles that are attached to them. When muscles are used, they incur slight amounts of damage, which causes the recruitment of inflammatory cells that remove debris and allow the native tissue to regenerate. When viewing muscles in cross section, muscle damage manifests as disruptions of normal muscle architecture, notably invaded muscle fibers and regions of densely packed cells. By quantifying this damage in a muscle, researchers may be able to more accurately assess the frequency with which that muscle is used and behaviors are displayed. Anoles have two muscles that attach to the hemipene, which is the intromittent organ used during copulation. Examining damage to the muscle that retracts the hemipene back into the tail (the retractor penis magnus (RPM)) may lead to better estimations of copulation rates in wild populations, which can be difficult to collect for species at low densities, or who mate cryptically.
Martin measured the muscle damage of the RPM for 5-10 males in 27 different species of anoles. This estimate was made by calculating the cross sectional area (CSA) of the RPM and the CSA area of muscle damage of each muscle. He then calculated a ratio of muscle damage (damaged CSA / total CSA) for each RPM and then averaged to give each animal a single value. These species were also observed in the field to measure an observed copulation rate (totaling ~1000 hours of observation). He and his collaborators used phylogenetical generalized least squares regressions to test for a correlation between observed copulation rates and the average ratio of muscle damage across these species.
Positive correlation between observed copulation rates and the average ratio of damage of the RPM for 28 species of anole.
They found a significant and strong positive correlation between these estimates, suggesting that examining muscle damage may be an efficient way to estimate behavioral rates. Martin drove home the point that measuring the damage in the RPM of these species took him 2 orders of magnitude less time than estimating copulation rates in the field. This suggests that researchers may be able to more easily estimate behavioral rates of different species, as well as examine individual variation within species. In the future, this group hopes to explore the relationship between muscle damage, copulation rate, and recovery so they can more accurately describe the window of behavior they observe through muscle damage.
The effects of climate change and urbanization on reptiles and amphibians has been a major topic at this year’s SICB. Both are expected to cause drastic changes in the climate, which will likely be severely detrimental to many species. We hope that many species will be able to tolerate these changes by adapting or acclimating, either by thermoregulating or changing their physiology. Adults of many species are able to acclimate in this way, but Josh Hall of the Warner lab wanted to test if eggs (which are unable to move to thermoregulate) are able to acclimate their physiology in response to higher average temperatures and to spikes in temperature that you might find in urban environments. Josh collected two populations of wild A. cristatellus from Miami, an urban population and a forest population, brought them back to the lab, and collected their eggs.
Josh Hall’s design, with 5 temperature regimes. City=light blue, Forest=green.
He then put the eggs into five different thermal conditions 1) higher “urban” temperatures, 2) cooler “forest” temperatures, 3) “urban” temperature with a large temperature spike on day 8, 4) “forest” temperature with a large spike on day 10, and 5) “forest” temperature with a small spike on day 10. He predicted that eggs that had a higher baseline temp (i.e. the urban eggs), would be able to tolerate spikes in temperature better than eggs at lower temperatures and would have higher survival and less of a physiological stress response due to the temperature spike. Contrary to his hypothesis, he found that high temperatures, and spikes were both detrimental to the survival of eggs and hatchlings, and affected baseline and max heart rate in embryos. This is somewhat concerning because it suggests that even a single short burst of heat can kill embryos, and have lasting affects on juveniles. The bursts, which you might expect in urban areas, have a more profound affect when the background temperature is higher, which will likely happen due to climate change.
Effects of incubation treatments on embryonic heartbeat, egg and juvenile survival.
The Johnson lab has another strong showing here at SICB 2017 with lots of presentations and posters! I stopped by two of their (many) posters on the evolution of reproductive behaviors and sexually-selected traits in anoles.
Adam Zeb, Amy Payne, and Hannah Hall presenting their posters at SICB 2017.
Adam Zeb and Amy Payne presented their poster that compared reproductive behaviors in anoles to the size of the neuromuscular junctions (NMJs) in the muscles responsible for dewlap extension (ceratohyoid) and hemipenes retraction (retractor penis). They predicted that species with higher dewlap extension rates and copulation rates would have larger NMJs because the NMJ is where the neuron communicates with the muscle fiber, initiating contraction. To ask this question they observed and collected 15 species of anoles from the Dominican Republic, Puerto Rico and the Bahamas. For each species they measured dewlap display rate, copulation rate, and also collected the ceratohyoid muscle and the retractor penis muscle. These tissues were sectioned and stained with acetylchloine iodide and lithium iodide to find the and measure the NMJs. This is still a work in progress, but preliminary evidence doesn’t suggest that NMJ area is correlated with retractor penis muscle size or with ceratohyoid size. However, there was a strong difference in NMJ size between those two muscle types, where the ceratohyoid has over 3x larger NMJs than the retractor penis muscle. This somewhat supports their original hypothesis that NMJ size would be correlated with use, as the dewlap is used much more frequently than the retractor penis muscle. Hopefully next SICB we’ll hear more!
Another Johnson lab member, Hannah Hall, has been working on a project to look at the relationship between pre- and postcopulatory traits in Anolis and to characterize the architecture of the Anolis testis. We know that Anolis have highly variable sperm morphology, but we do not know if a portion of that variation may be due to variation in the structure of the testis. To test this, Hannah collected testis sections from eight species of anoles, and measured the cross-sectional area of the seminiferous tubule, the lumen and the epithelial height. She also collected measurements of sexual size dimorphism (as a proxy for the strength of precopulatory selection) and gonadosomatic index (GSI), which is the ratio of testis mass and body mass (as a proxy of the strength of postcopulatory selection). She found a negative correlation between SSD and GSI, suggesting a trade-off between pre- and postcopulatory selection. She also found significant positive correlations between cross-sectional area of the testis and sperm head size, and between lumen size and sperm tail size. This suggests that larger structures in the testis are needed to produce sperm with larger morphology. Hannah is still working on characterizing the testis structure of many anole species, so stay tuned for more developments!
Many anoles have prolonged breeding seasons spanning from the late spring until the early fall. For part of his Master’s degree Phillip Pearson, a student in the Warner lab at Auburn University, asked whether the timing of oviposition is adaptively matched to a season’s thermal environment and if there are fitness consequences of early or late developmental temperatures in Anolis sagrei. They predicted that eggs laid early in the season (April-May) and were incubated under ‘early season’ temperatures would have higher hatchling fitness than under ‘late season’ (July-August) temperatures, and that late-produced eggs would have higher fitness in ‘late season’ temperatures than ‘early season’ temperatures.
To test this hypothesis Phillip collected adult males and females from the wild and brought them back to the lab to breed. He then collected eggs from March-April as the ‘early’ cohort and from July-August as the ‘late’ cohort. Each of these cohorts was then divided into two treatments with ‘early season’ and ‘late season’ incubation temperatures, resulting in four groups. Each hatchling was weighed, measured and assessed for sprint performance.
Phillip found that both the time of oviposition and the incubation temperature significantly affected the development of the hatchlings in several ways. First, eggs in both the early and late cohorts that were incubated under early temperatures had significantly longer incubation durations. Temperature also interacted with the season cohorts, so that the ‘late season’ cohort incubated under the late season temperatures had the shortest incubation duration (Figure 1A). Second, Phillip found a significant effect of season, incubation temperature and their interaction on egg survival, where the late season cohort that was incubated under late season temperatures had the highest survival (Figure 1B). However, he did not find a significant effect of either incubation temperature or season cohort on hatchling survival. Third, eggs that were laid in the late season cohort were significantly larger in mass, snout-vent length, and tail length at hatching than early-season eggs (Figure 1C). Finally, hatchlings from the ‘late season’ cohort had marginally faster sprint speeds, with more stops (Figure 1D).
A. Incubation duration, B. egg survival, C. hatchling mass, and D. sprint speed for eggs oviposited in ‘early season’ and ‘late season’ cohorts raised under two different incubation temperatures.
Overall, Phillip’s results suggest that eggs laid later in the season and incubated under warmer late-season temperatures seem to have higher performance and fitness (in some cases). Currently, Phillip has released these hatchlings onto an island in Florida near the site of the parent population. He and the Warner lab will be going back this spring to assess survival of these hatchlings to get field-relevant data on survivorship under these two developmental treatments.
Muscle and jaw morphology is highly variable among lizards, which could be driven, at least in, part by a species’ diet, intraspecific combat, or both. Leah Selznick of the Johnson lab at Trinity University collected data on the head dimensions, jaw muscle mass, diet data (prey count), and estimates of sexual dimorphism (SSD) for seven species of lizards. Four of the species she examined – the leopard gecko, Northern curly tail, Texas spiny lizard, and the Carolina green anole – are saurophagous, meaning that they eat other lizards. The other three species – the Mediterranean house gecko, little brown skink, and spotted whiptail – do not eat other lizards. Leah predicted that saurophagous species and those with a higher variance in prey diet would also have larger heads and larger jaw muscles. Additionally, she predicted that species with higher sexual size dimorphism (SSD), a proxy for the strength of pre-copulatory selection on male body size, would be associated with larger jaw morphologies. First, she tested for phylogenetic signal in all of her traits and found strong signal for jaw muscle mass (λ = 0.99) and head size (λ = 0.65). She then tested for an association between both head size and jaw muscle mass (standardized by body mass) with species prey count and SSD. She found no correlation between any of the jaw morphologies and SSD or species prey count. Leah suggests that (1) there may be other traits that are experiencing selection due to prey size and combat and/or that (2) these traits may be experiencing evolutionary constraint. Leah is going to continue exploring the evolution of jaw morphology by examining the histology of the jaw muscle in these species to test for an association between muscle fiber composition and type with prey count and SSD.
Selznick compared two groups of lizard species: four saurophagous species, and three exclusively insectivorous species. Here, she shows the prey count, and SSD for each species used in her analysis.
When I was first designing projects for my dissertation, a result from one of my advisor’s papers caught my attention – brown anole males in better body condition (relatively more massive for their body size) sired more offspring and more sons. We didn’t have an explanation for how or why this trend existed but as a wannabe sperm biologist, I was immediately suspicious that it had something to do with sperm quality. I had some preliminary data showing that brown anole males varied in their sperm morphology and sperm count, but I wanted to know if some of this intraspecific variation was due to condition dependence and if there were fitness consequences associated with this variation.
Male brown anole in St. Augustine, FL.
In our recent experiment, we tested whether body condition was correlated with sperm quantity and quality, and whether the variation in sperm traits resulted in differences in a male’s competitive ability. To do this, we placed two groups of males on high-intake and low-intake diet treatments, where males were fed either five crickets three times a week or one cricket three times a week to experimentally alter their body condition. They were fed this diet until the two groups diverged in condition, and then kept on the diet treatments long enough for them to develop a fresh batch of sperm while in this altered body condition. We collected a sperm sample and measured sperm count and the morphology of 25 cells for each male. We focused on measuring the three largest regions of the cell, the head, the midpiece and the tail (see image below). To test for differences in the ‘competitiveness’ of each group’s sperm, we designed reciprocal mating trials so that a pair of males (one male from each group) would compete for fertilization of a female’s brood. Each male pair was mated to two females, and the order in which the males mated with the female was reversed for the second female to account for mate order effects.
Figure 2 from Kahrl and Cox 2015, (A). Anolis sagrei sperm cell B. Individual means (±SD) for head length, midpiece length, and tail length of 25 sperm cells per individual for each of 17 males from each treatment group (high- and low-intake). (C) Treatment means (± standard error) of individual means in head length, midpiece length, and tail length. (D) Treatment means (±SE) of individual CV in head length, midpiece length, and tail length.
To complement this lab study, we collected sperm from a wild population of brown anoles to look for condition dependence of sperm morphology in the wild. We also reanalyzed paternity data from Cox et al. 2011 to test for condition-dependent reproduction in a lab population of brown anoles. It should be noted that the lab population in this study (Cox et al. 2011) differed from our experimental population in a few ways. First, the males from that study did not have experimentally manipulated body condition. They were all fed the same diets, and the pairs of males that contained both a male in naturally high-condition and low-condition were included in this analysis. Secondly, though the mating design in that study was the same as our experimental reciprocal design, in Cox et al. 2011 males were allowed unlimited access to the females for an entire week, where in our experimental study males were limited to a single copulation.
Figure 4 of Kahrl and Cox 2015. Mean (± standard error) proportion of progeny sired by males that were (A) categorized into high- and low-condition pairs (data reanalyzed from Cox et al. 2011) and (B) assigned to high-intake and low-intake diet treatments. Condition dependence was assessed in 3 ways: 1) using each dam as a unit of observation and estimating the proportion of paternity for each of her 2 mates, 2) using each pair of potential sires as a unit of observation and estimating the proportion of paternity for each male, and 3) using each pair of potential sires as a unit of observation but restricting the comparison to the subset of pairs for which both dams produced offspring.
We found that in both the lab and field, males in low body condition or on a low-intake diet treatment had significantly larger and more variable sperm midpieces than males in high body condition. We also found that males on the low-intake diet treatment had significantly lower sperm counts. When we analyzed the paternity data to test for correlations between fertilization success and sperm traits, we found significant negative correlations between sperm head and midpiece length, sperm count and fertilization success (though it should be noted that we only found these correlations for the average proportion of paternity and not when males were analyzed by either the proportion of paternity from their first or their second mating). We tested for condition-dependent fertilization success in our experimentally manipulated population and reanalyzed the data from males who varied naturally in body condition from Cox et al. 2011. We found a significant difference in fertilization success in males who varied naturally in body condition and had unlimited access to females, but found no difference in fertilization success in males who were in the experimental diet treatment groups (though the trend was similar in our experiment). Together, these data suggest that condition-dependent fertilization success is partially mediated by sperm quantity and morphology, and may also be influenced by a male’s ability to mate multiply with the same female.
This is the first paper that is part of my dissertation on the evolution of sperm morphology. I’m using anoles and phrynosomatid lizards to assess the sources and consequences of inter- and intraspecific variation in sperm morphology. Hopefully I’ll have more to share about anole sperm biology soon!
Kahrl, A.F., and R.M. Cox. 2015. Diet affects ejaculate traits in a lizard with condition-dependent fertilization success. Behavioral Ecology (advance access).
Jake Stercula of the Johnson Lab and his 2015 SICB poster
Muscle fiber size can vary based on the frequency of use, or due to the fusion of multiple mononucleated myoblasts during development to form multinucleated fibers. To test if variation in muscle fiber size was due to frequent use or due to differences in development between species, Jacob Stercula of the Johnson lab examined the fiber size and number of nuclei for the ceratohyoid and the retractor penis magnus (RPM) of nine species of anoles. Most species exhibit a positive relationship between fiber size and the number of nuclei in both muscle types. Among species, this positive relationship between fiber size and the number of nuclei exists in the RPM muscle when accounting for phylogeny using independent contrasts, whereas the ceratohyoid shows a positive trend, though the relationship was not significant. This suggests that for the RPM, muscle fiber size is evolutionarily conserved and is due to differences in development among species rather than differences in the amount of use. The size of the ceratohyoid muscle however, maybe be influenced by both the frequency of use and the fusion of myoblasts during development.
Michelle Oberndorf of the Johnson Lab (Photo courtesy of the Johnson Lab website)
Anolis visual display can come in two flavors: static and dynamic. Static displays are those that are involve permanent morphological structures, whereas dynamic displays involve movement of physical structures. Michelle Oberndorf of the Johnson lab asked if structures involved in both static display (tail crest) and dynamic display (dewlap size), were related to body condition or fighting ability (head size) in males and females of two species of anoles. She collected SVL, mass, head morphology, tail crest size, and dewlap size data from 50 males and 50 females of A. cristatellus and A. gundlachi. She found that in males, tail crest area was correlated with body condition in A. cristatellus. In male A. gundlachi, tail crest area was correlated with head size, and dewlap size, while dewlap size was correlated with body condition and head size. She found no relationships between any of the traits in females of either species. These results suggest that both the dewlap and the tail crest may communicate information about male quality and potential fighting ability.
Correlations between morphology and body condition for A. cristatellus and A. gundlachi males. Image from Michelle’s poster.
