SLIDE 1: Title Slide

TSD in Reptiles: Modes, Mechanisms and Evolution.

SLIDE 2: Sex modes in reptiles

In contrast to mammals and birds, reptiles have an impressive array of sex determining modes, comparable to the variety observed in fish and frogs. Even among the lizards, we have

·  Male Heterogamety

·  Female Heterogamety

·  GSD with no Heterogamety

·  Parthenogenesis

·  Temperature-dependent Sex determination

SLIDE 3: Definitions

In species with GSD, sex is considered to be determined by genetic factors operating at the time of conception, and they operate largely independently of environmental influences.

In species with TSD, sex is determined after fertilization by the environmental conditions that prevail during embryonic incubation, and it is determined largely independent of direct genetic influences.

The two modes of sex determination are often viewed as fundamentally different. It has been thought that the differences in the mechanisms between the two modes are complex, that they constitute a discrete dichotomous process, and that through appropriate experimental approaches, one can be demonstrated to the exclusion of the other.

SLIDE 4: Objectives of the talk

In this talk, I will

·  Present an overview of TSD in reptiles

·  Argue that TSD and GSD in reptiles are best viewed ends of a continuum, not a dichotomy, where genetic and environmental influences may coexist and interact in greater or lesser degree in different species to bring about sexual phenotypes.

·  Speculate about the possibility that reptiles are, in a sense, predisposed to TSD.

·  On the way, I will present a little bit about our work and where it is heading, linking across to the later presentation by Tariq Ezaz.

SLIDE 5: Early work

TSD was first discovered by Charnier in the 1966, when he found that incubating eggs of Agama agama at low temperatures produced predominantly one sex (F:M 45:1). This finding has been confirmed for a wide range of reptiles, including:

·  Many other lizards across many families, but not snakes, and most recently including the viviparous Eulamprus tympanum.

·  All species of Crocodilians

·  All species of marine turtle, and many species of freshwater turtle

·  The tuatara

SLIDE 6: Details of the effect

The environmental effect can be quite dramatic. In our work on the pignosed turtle, for example, you have 100% males produced at low temperatures, 100% females produced at high temperatures. A very narrow range of temperatures referred to as the pivotal temperature or pivotal range (SLIDE 6B) produces both males and females. Half a degree in one direction or the other can make all the difference.

In other species, such as the Tuatara and the viviparous Eulamprus tympanum the reverse pattern is true, with males produced at high temperatures and females produced at low temperatures.

SLIDE 7: Loggerhead turtle

The pivotal range need not be narrow, and differs in position and width among species. In the loggerhead turtle, the pivotal temperature is 28.5C (compared to 32C for the pignosed turtle) and the pivotal range extends across 4.5C. It is a ramp function.

SLIDE 8A: Water Dragon

Some turtles, crocodiles and many lizards produce males only at intermediate temperatures, and females at both extremes. Females are commonly produced at all temperatures, with only a strong male bias at intermediate temperatures. Two pivotal temperatures are identified (SLIDE 8B), though often only one occurs within the range of temperatures experienced in the field.

SLIDE 9: Underlying pattern

It is believed that there is a common underlying pattern for all reptiles, the FMF pattern, which is expressed as an MF or FM pattern when the conditions suitable for egg survivorship are overlaid. So the underlying mechanisms might well be common across the varying patterns we see in nature.

SLIDE 10: Switch Experiments

Experiments where eggs are switched between male and female producing domains at strategic points during incubation have shown that the middle third of incubation is when temperature exerts an irreversible effect on offspring sex. This seems to be a common feature across all groups, and the period coincides with the differentiation of the gonad into male and female.

Interestingly, there is an accumulation effect as switching for only two embryonic stages at any point in incubation does not switch sex, whereas switches of 3 stage duration does.

Switching from low to high temperatures and back again seems to be more influential than switching from high to low temperatures and back again.

SLIDE 11A: Variability

Of course reptile nests in the field do not obey the constraints we place on eggs in the laboratory. Daily fluctuations in temperature, seasonal trends and stochastic events such as rainfall, which temporarily depress nest temperatures, can all be expected to influence sexual outcomes. And they do.

We have incubated eggs of the loggerhead turtle at a mean temperature of 26C, which fits squarely in male producing domain, but varied the daily fluctuations about that mean to varying degrees. By simply changing the magnitude of the daily fluctuations, we have been able to switch sex from 100% male to 100% female. This is without changing the mean temperature, or for that matter the incubation period, at all.

We have modeled this under assumptions that it is the amount of development that occurs at a temperature, not the duration of exposure, that is influential in determining sex, and obtained very good agreement between the model and observation (SLIDE 11B).

SLIDE 12: Are the two modes fundamentally different?

Are the two modes fundamentally different? In a recent paper, we have argued that they are not, citing as evidence the fact that

·  There is remarkable conservatism of vertebrate sexual differentiation genes, including reptiles with TSD.

With the notable exception of SRY, many genes involved in gonadal differentiation in mammals (SF1, DMRT1, SOX9, AMH, DAX1 and WT-1) have homologues in reptiles. Of these, DMRT1, SOX9, SF1, DAX1, AMH and WT-1 are expressed during gonadogenesis in alligators, and DMRT1, SOX9 and WT-1 are expressed during gonadogenesis in turtles. Several, including SF1, SOX9 and AMH, are expressed differentially in males and females reptiles with TSD.

·  Taxonomic distribution of TSD and GSD species suggests both modes have evolved independently multiple times.

The distribution of TSD across the reptiles appears almost haphazard. At the level of order, we have TSD in all crocodiles, yet their sister taxon the birds universally has GSD. At the level of family, we have the monotypic family that includes the pignose turtle with TSD, but its sister taxon the Tryonichidae has GSD. This pattern is reflected in the Chelidae/Pelomedusidae pair. At the level of genus, we have species with GSD, and other species in the same genus with TSD – such as in the Clemmys group of Emydid turtles and the Amphibolurus group of Agamid lizards.

It appears that the line between TSD and GSD has been crossed on many occasions in the evolutionary history of reptiles, probably in both directions. This suggests that the evolutionary transitions between the two mechanisms may be relatively simple, perhaps involving one or only a few genes.

SLIDE 13: Where is temperature sensitivity achieved

I do not want you to get too excited about this slide. It is strictly hypothetical to make a point. The genes shown are those known to occur in reptiles, present also in mammals or birds, where in at least one these groups they have been shown to cause sex reversal. There arrangement follows that of Pieau and Wilkins, with some poetic licence.

The point is that temperature sensitivity could be achieved by a mutation at many points in the differentiation cascade, not just a mutation on the complex responsible for sexual determination.

·  Temperature sensitivity might occur through a mutation in the top sex determining gene or gene complex.

·  Temperature sensitivity could conceivably occur in the genes governing the environment where the sex determining products act (the action of SRY protein depends on the function of other genes and products that influence its passage through the nuclear membrane).

·  Temperature sensitivity could occur in some other gene high in the sexual differentiation cascade. Wilkins has noted increasing conservatism across taxa as you move down the cascade, so changes higher in the cascade might be more likely.

·  Temperature sensitivity could occur in the aromatase gene, or in genes influential in its expression, as manipulating aromatase will cause sex reversal in reptiles.

·  Temperature sensitivity could occur at the level of the oestrogen receptors, affecting the efficacy with which estrone and estradiol do their work.

Interesting questions arise.

·  Is there a predisposition for one part of the sex determining/sex differentiation cascade to become temperatures sensitive, or has it occurred multiple times and in multiple different ways across reptile taxa. We might revisit that question later.

·  Once temperature sensitivity has been achieved, how much of the (redundant) sex determining machinery higher in the cascade is retained, and if it is retained, does it continue to have some influence.

SLIDE 14: Dichotomy or Continuum

That is, do genetic and environmental influences coexist in some taxa, interacting in greater or lesser extent to bring about sexual outcomes. Do our current notions of TSD and GSD represent extremes in a continuum, with many species residing at intermediate positions?

SLIDE 15A: H-Y antigen evidence

Good evidence for the coexistence of genetic and environmental influences on sex determination comes from the studies of Pieau and his colleagues.

Emys orbicularis has temperature dependent sex determination. Pieau and his colleagues found that male offspring of Emys orbicularis produced by incubating at low temperatures had H-Y antigen expression in the gonads that was concordant with phenotypic sex, but that in the blood, expression was 50:50. A similar result was obtained for females produced at high incubation temperatures (SLIDE 15B).

When incubated at the pivotal temperature, however, HY antigen expression in the blood concurred with phenotypic sex (SLIDE 15C). This is strong evidence of an underlying genetic influence on sex, over-ridden by temperature at all temperatures but those for which the temperature effect is ambiguous – at the pivotal temperature.

Interestingly, the incidence of discordance between genotypic predisposition and phenotypic sex in field nests was not high (SLIDE 15D). This suggests that the potential temperature over-ride does not often occur under natural conditions in this species.

SLIDE 19: Bassiana

Even stronger evidence for the possible coexistence of genetic and environmental influences comes from studies of a small lizard in the Australian high country. Bassiana has heteromorphic sex chromosomes, a trait possessed without variation across the genus, yet temperature has been shown to have an impact on offspring sex. At intermediate temperatures, sex is determined genetically in the usual way, but at low temperatures, this genetic influence gives way to temperature.

SLIDE 20: Pogona

In our own lab, we have been working with the bearded dragon. It has homomorphic sex chromosomes, like all agamids, but it regarded as a GSD species. We have found that sex ratios are unaffected by temperature at all but the upper extreme of 34C. At 34C, there is a strong female bias. These results should be regarded as preliminary, and we will be pushing the limits both high and low in coming months, but it is suggestive of coexistence of genetic and environmental influences in this species also.

We will be looking for sex specific markers in this species to see if they assort with phenotypic sex at all but the extreme temperatures.

SLIDE 21: Continuum

So are the mechanisms of TSD and GSD best viewed as a dichotomy? We would say no, and cite in support the

•  Extraordinary conservatism in the genes involved in sexual differentiation across vertebrates

•  Abundant examples in fish of co-existence of genetic and environmental influences

•  Evidence of co-occurrence of genetic and environmental influences in at least some reptiles

•  Relative ease by which species appear to have moved from one mode to the other in evolutionary history, suggesting only relatively minor changes are required to do so.

SLIDE 22: Sister taxa

In this context, we believe that comparative study of closely related taxa, one with TSD, the other with GSD, provides the best approach.

We have selected two species of Amphibolurus, very closely related, and undertaking cytological and molecular comparisons to gain insight into the differences between the two. Tariq will talk more about our approaches, but in particular, we will be seeking sex specific markers in the GSD species and seeing if they are present and assort 50:50 in the TSD species (though not assorting with phenotypic sex). We will be looking also to see if phenotypic sex and the assortment of the marker concur at the pivotal temperatures in the TSD species.

This will open up a wide range of possibilities for field studies of the selective pressures that operate to maintain TSD in natural populations – what are the implications of concordance in sex (genotype=phenotype) versus discordance (genotype≠phenotype) in terms of fitness and contribution to future generations.

A fruitful future I think, if we can get a system of molecular markers to distinguish between underlying "genotypic sexes" in TSD species.

SLIDE 23: Other side of the coin

We have looked at TSD species and presented evidence that, in at least some species, genetic and environmental influences on sex co-exist and interact to produce sexual phenotypes. I would now like briefly to look at the other side of the coin. Do GSD reptiles have an underlying predisposition to evolve TSD?

Is it a coincidence that both homeothermic classes have sex determined universally by genetic sex determination?

Is it a coincidence that reptiles, whose embryos experience the widest range of temperatures during incubation of any vertebrate group (22-45C within a single nest within a single day), have the greatest prevalence of TSD?

SLIDE 20:

It would seem necessary for reptiles whose embryos experience a wide range of temperatures, either as individuals, or across nests, that there exists a mechanism that is resistant to temperature change.

How otherwise could a consistent and unambiguous signal be sent from the sex determining gene or gene complex and the hormonal environment that is proximately responsible for sexual differentiation.