Polymorphic male color morphs visualized with steroids in monomorphic females: a tool for designing analysis of sex-limited trait inheritance

Many organisms show variation within distinct categories, such as castes in termites and color morphs in lizards (Sinervo and Lively, 1996). To understand their evolution, researchers benefit from designing explicit breeding experiments that elucidate their patterns of inheritance so that theoretical models can be used to predict their response to selection in the wild (Lynch and Walsh, 1998). This is confounded when morph-specific traits such as color are expressed in only one sex, as it is unknown which color category one parent genetically codes for and therefore what color allele will be transferred to offspring by the colorless parent. These morphs may also encompass suites of coevolved morph characteristics from the genomic to the physiological, reproductive and ultimately evolutionary level (Sinervo and Lively, 1996). However, because the most overt part of the phenotype is only observable in one sex, one may overlook the fact that parts of the coevolved genome and trait repertoire can have profound effects on the biology of the sex in which the morph-specific characteristics are never seen (Lank et al., 1999). In order to be able to study this, the phenotypic morph characteristics (e.g. color) can be revealed experimentally.

Our model of choice for this work was the Australian painted dragon lizard (Ctenophorus pictus, Peters 1866). Males of this species occur in three different color morphs (red, orange and yellow headed), whilst females are cryptic in coloration and both males and females may express yellow pigmentation on their throats [a ‘bib’ (Healey and Olsson, 2009; Olsson et al., 2009a)]. Red males are the dominant morph in staged contests, facilitated by higher levels of testosterone, aggression and willingness to fight. They also emerge from hibernation sooner (Healey et al., 2007). Yellow males counter this aggressive tactic via sperm competition, siring more offspring per copulation than red males (Olsson et al., 2009b). Orange males are characteristically close to the red morphotype with respect to morph-specific traits (Healey and Olsson, 2009). We therefore asked whether the different, morph-specific strategies between males in terms of levels of aggression are also reflected at a metabolic cost level – specifically, in their levels of reactive oxygen species (ROS) and the antioxidant superoxide dismutase (SOD). This also raises the fascinating question of whether this morph specificity also applies to females, despite the striking differences in male and female lifestyles. Although females are non-territorial and non-aggressive, they do still carry the genetic underpinning of a specific morph and any co-variation between head color, ROS and SOD levels in males could also therefore be present in females.

The Australian painted dragon, C. pictus, is a small (adult snout–vent length 65–95 mm, mass 8–16 g) diurnal lizard. They typically prefer sandy habitats and low vegetation in a range covering central and western New South Wales to Western Australia. Orange and yellow male lizards, and the monomorphic females, were caught by noose at Yathong Nature Reserve, New South Wales (145°35′E, 32°35′S) during the 2009 mating season (August–January), weighed, measured (only relevant measures reported) and scored for head coloration and bib presence/absence by eye (see Sinervo and Lively, 1996). Red males did not occur in high enough frequency in 2009 to be included in the experiment. The lizards were then brought back to holding facilities at Wollongong University where they were held separately in 600×600×500 mm cages with a 40 W spot light at one end to allow thermoregulation to their preferred body temperature (ca. 36–37°C as per cloacal temperatures of free-ranging territorial, displaying males; M.O. and E. Wapstra, unpublished). Lizards were fed crickets and mealworms dusted with calcium and multivitamins ad libitum every second day and sprayed with a mist of water twice daily. Lizards were allowed to acclimatize to captive conditions for 2 weeks, after which time they were blood sampled for flow cytometry and superoxide quantification.

A single sample of peripheral blood (25 μl) was drawn in the morning from the vena angularis (corner of the mouth) using a syringe to puncture the blood vessel and a capillary to collect the blood. The blood was diluted immediately with 9 volumes of phosphate-buffered saline (PBS; 137 mmol l^–1 NaCl, 2.7 mmol l^–1 KCl, 1.5 mmol l^–1 KH₂PO₄, 8 mmol l^–1 Na₂HPO₄, pH 7.4) and stored on ice prior to analyses, which were completed within 4 h of sampling. Prior to staining, diluted blood was further diluted 50-fold with PBS and then centrifuged (300 g for 5 min) to pellet cells; each cell pellet corresponded to 10 μl of whole blood. Cells were re-suspended in 100 μl of PBS containing 5 μmol l^–1 MitoSOX Red (MR; Molecular Probes, Invitrogen, Carlsbad, CA, USA). MR was added from a stock solution in dimethylsulfoxide (DMSO). The final concentration of DMSO was 0.2% (v/v) or less. Cells were subsequently incubated at 37°C for 30 min, then washed with PBS by centrifugation as described above and held on ice until analysed by flow cytometry; 50,000 events were acquired for all samples. Flow cytometry was performed using a Becton Dickinson LSR II (Becton Dickinson, Sydney, Australia), with excitation at 488 nm and emitted fluorescence collected using band pass filters of 515±10 nm (DHR, dihydrorhodamine 123, Molecular Probes) and 575±13 nm (MR). Data were acquired and analysed using FACSDiva v4.0.1 and FloJo software (Becton Dickinson), respectively. On the basis of forward angle laser scatter and side angle laser scatter, a number of blood cell populations were discerned. The results obtained were similar for all these populations. For each sample, the arithmetic mean fluorescence for all 50,000 cells acquired was determined using FloJo software and used to compare samples and treatments. The accuracy of flow cytometry results from two samples from the same individual has been measured in a separate experiment (Olsson et al., 2008), involving 14 males with a correlation coefficient between samples of r=0.97, P<0.0001. Thus, our flow cytometry technique can be argued to be highly consistent.

As SOD scavenges superoxide molecules, we assayed superoxide and entered those values as a covariate in our analysis. We used Superoxide Dismutase Assay Kit II (cat. no. 574601, Calbiochem, supplied by Merck Pty Ltd, Victoria, Australia). This kit utilizes a tetrazolium salt for detection of superoxide radicals generated by xanthine oxidase and hypoxanthine. One unit of SOD is defined as the amount of enzyme needed to exhibit 50% dismutation of the superoxide radical. The SOD assay measures all three types of SOD (Cu/Zn-, Mn- and Fe-SOD).

After all animals had been characterized with respect to superoxide and SOD, and all females had laid their clutches for the season, the females (only) were implanted with testosterone to express their underlying color. Females were sterilized dorsally by wiping with 70% alcohol and injected subcutaneously with 0.2 ml lidocaine (lidocaine hydrochloride 0.4%; Sigma Aldrich Pty Ltd, Castle Hill, NSW, Australia). They were then cooled in a sterile plastic box on ice (being ectotherms, this immobilizes lizards without causing any harm) and wiped ventrally with 70% alcohol. A small incision was made between two scale rows and sterile implantation of hormone took place; subsequently, the incision was closed with surgical superglue (2-octyl cyanoacrylate; Hystoacryl). The 8 mm Silastic implant was made of a Dow Corning tube (i.d. 1.5 mm, o.d. 2.0 mm; Dow Corning, Midland, MI, USA). Testosterone crystals were packed into a 6 mm column (no. T 1500, Sigma) and one end of the implant was closed with Dow Corning 734 RTV sealant. Implanted females began to express head color within a week and after 4 weeks were brightly colored, making the assessment of the color morph by eye straightforward.

We initially used a three-factor ANOVA to assess the effect of sex, head color (orange and yellow) and bib (present, absent), while including two covariates, superoxide (controlling for among-individual variation in superoxide) and individual SOD levels. We then backward-eliminated the non-significant three-way interaction term.

Two of our interaction terms were significant: between head color and bib presence, and between sex and bib presence (Figs 1, 2, 3, Table 1). However, the interaction between sex and head color was non-significant, suggesting that different head colors have different superoxide levels, but with no difference between the sexes. Thus, in spite of females not showing the colors they genetically code for, they do show similar genetic morph net levels of superoxide to males in the corresponding morphs. Sex itself had a non-significant effect on superoxide (P=0.083), while both bib and head color showed significant main effects (P=0.007 and P=0.013, respectively). However, as these also interacted significantly, this casts doubt on their significant main effects.

We use testosterone treatment to reveal the otherwise ‘silent’ head color in females, allowing us to compare morph and sex-specific levels of superoxide and show that even though females do not express the phenotypic colors of males, their net levels of superoxide as a result of oxidative metabolism and antioxidation follow the same genetic morph specificities as those in males. Furthermore, additional variation in superoxide was explained by the presence or absence of a throat bib within head color morphs. Interestingly, females with bibs have less superoxide than those without bibs, and this pattern was not evident in males. What could explain these similarities and differences in a major metabolic parameter (net superoxide level) between the sexes? At least two different routes seem possible where genetic morph specificity in ROS levels could arise, regardless of sex and whether a phenotypic color is expressed or not. (i) Animals may behave the same with respect to some external ROS moderator. This could arise via selection on color morphs and coevolved traits in one sex (males) where only some of these traits become expressed as a result of genetic covariation in the other sex. (ii) Alternatively, there may be physiological or biochemical differences that are first under diversifying selection in both sexes, and any behavioral differences associated with them become selected for in a second stage of morph divergence or early speciation process (Sinervo and Lively, 1996). Analysis of these alternative scenarios is outside the scope of this study.

In summary, we demonstrate that exploiting hormonal manipulation to reveal hidden morphs in the monomorphic sex gives us a tool to quantify physiological and biochemical parameters involved in the evolution of polymorphism in both sexes.

We thank E. Snaith and G. Snaith for logistical support.