Temperature-dependent sex determination

Temperature-dependent sex determination (TSD) is a type of environmental sex determination in which the temperatures experienced during embryonic development determine the sex of the offspring. It is most prevalent and common among amniote vertebrates that are classified under the reptile class, but is absent among birds, including the Australian Brush-turkey, which was formerly thought to exhibit this phenomenon. TSD differs from the chromosomal sex-determination systems common among vertebrates. It is a type of environmental sex determination (ESD); in other ESD systems, some factors such as population determine the sex of organisms (see Polyphenism).

The eggs are affected by the temperature at which they are incubated during the middle one-third of embryonic development. This critical period of incubation is known as the thermosensitive period (TSP). The specific time of sex-commitment is known due to several authors resolving histological chronology of sex differentiation in the gonads of turtles with TSD.

Thermosensitive Period (TSP)
The thermosensitive, or temperature-sensitive, period (TSP) is the point during development at which sex determination occurs in temperature-dependent sex determination systems, such as those present in alligators and turtles. Occurring during the middle third of incubation, the TSP lasts for 7-15 days, species-dependent. The temperature must be maintained during this time for a certain sex to be determined; often times the differences between temperatures required to produce each sex are very slight and during the TSP, sex is susceptible to change with temperature. After this period, however, sex cannot be reversed.

Types
Within the mechanism, two distinct patterns have been discovered and named Pattern I and Pattern II, with Pattern I further divided into IA and IB. Pattern IA has a single transition zone, where eggs predominantly hatch males if incubated below this temperature zone, and predominantly hatch females if incubated above it. Pattern IB also has a single transition zone, but females are produced below it and males above it. Pattern II has two transition zones, with males dominating at intermediate temperatures and females dominating at both extremes. Very near or at the pivotal temperature of sex determination, mixed sex ratios and, more rarely, intersex individuals are produced.

In turtles with TSD, males are generally produced at lower incubation temperatures than females (TSD IA), with this change occurring over a range of temperatures as little as 1-2 °C. At cooler temperatures ranging between 22.5 and 27 degrees Celsius mostly male turtles arise, and at warmer temperatures around 30 degrees Celsius only female turtles arise. In lizards and crocodilians, this pattern is reversed (TSD IB).

It has been proposed that essentially all modes of TSD are actually type II and those that deviate from the expected female-male-female pattern are simply never exposed to extreme temperature ranges on either end.

Hormones in TSD systems
Synergism between temperature and hormones has also been identified in these systems. Administering estradiol at male-producing temperatures generates females that are physiologically identical to temperature-produced females. The reverse experiment, males produced at female temperatures, only occurs when a nonaromatizable testosterone or an aromatase inhibitor is administered, indicating that the enzyme responsible for conversion of testosterone to estradiol, aromatase, plays a role in female development. Nonetheless, the mechanisms for TSD are still relatively unknown, but in some ways, TSD resembles genetic sex determination (GSD), particulary in regards to the effects of aromatase in each process. <ref. . In some fish species, aromatase is in both the ovaries of female organisms who underwent TSD and those who underwent GSD, with no less than 85% of the coding sequences of each aromatase being identical, showing that aromatase is not unique to TSD and suggesting that there must be another factor in addition to it that is also affecting TSD.

Interestingly, hormones and temperature show signs of acting in the same pathway, in that less hormone is required to produce a sexual shift as the incubation conditions near the pivotal temperature. It has been proposed that temperature acts on genes coding for such steroidogenic enzymes, and testing of homologous GSD pathways has provided a genic starting point. Yet, the genetic sexual determination pathway in TSD turtles is poorly understood and the controlling mechanism for male or female commitment has not been identified.

While sex hormones have been observed to be influenced by temperature, thus potentially altering sexual phenotypes, specific genes in the gonadal differentiation pathway display temperature influenced expression. In some species such important sex-determining genes as DMRT1 and those involved in the Wnt signalling pathway could potentially be implicated as genes which provide a mechanism (opening the door for selective forces) for the evolutionary development of TSD. While aromatase is involved in more processes than only TSD,it has also been shown to play a role in certain tumor development<ref. .

Adaptive significance
The adaptive significance of TSD is currently not well understood. One possible explanation that TSD is common in amniotes is phylogenetic inertia – TSD is the ancestral condition in this clade and is simply maintained in extant lineages because it is currently adaptively neutral or nearly so. Indeed, recent phylogenetic comparative analyses imply a single origin for TSD in most amniotes around 300 million years, with several more recent independent origins of TSD in squamates and turtles. Consequently, the adaptive significance of TSD in all but the most recent origins of TSD may have been obscured by the passage of deep time, with TSD potentially being maintained in many amniote clades simply because it ‘works’ (i.e. has no overall fitness costs along the lines of the phylogenetic inertia explanation).

Other work centers on a 1977 theoretical model (the Charnov–Bull model), predicted that selection should favour TSD over chromosome-based systems when "the developmental environment differentially influences male versus female fitness"; this theoretical model was empirically validated thirty years later but the generality of this hypothesis in reptiles is questioned. This hypothesis is supported by the persistence of TSD in certain populations of spotted skink (Niveoscincus ocellatus), where it is advantageous to have females early in the season. The warmth early in the season ensures female-biased broods that then have more time to grow and reach maturity and possibly reproduce before they experience their first winter, thereby increasing fitness of the individual.

In support of the Charnov and Bull hypothesis, Warner and Shine (2008) showed confidently that incubation temperature influences males’ reproductive success differently than females in Jacky Dragon lizards (Amphibolurus muricatus) by treating the eggs with chemicals that interfere with steroid hormone biosynthesis. These chemicals block the conversion of testosterone to oestradiol during development so each sex offspring can be produce at all temperatures. They found that hatching temperatures that naturally produce each sex maximized fitness of each sex, which provides the substantial empirical evidence in support of the Charnov & Bull model for reptiles.

An alternative hypothesis of adaptive significance was proposed by Bulmer and Bull in 1982 and supported by the work of Pen (2010). They conjectured that disruptive selection produced by variation in the environment could result in an evolutionary transition from ESD to GSD (Bull, Vogt, and Bulmer, 1982). Pen (2010) addresses evolutionary divergence in SDM’s via natural selection on sex ratios. Studying the spotted skink, a small lizard in Tasmania, he observed that the highland population was not affected by temperature, yet there was a negative correlation between annual temperature and cohort sex ratios in the lowlands. The highlands are colder with a higher magnitude of annual temperature fluctuation and a shorter activity season, delaying maturity, thus GSD is favored so sex ratios are not skewed. However, in the lowlands, temperatures are more constant and a longer activity season allows for favorable conditions for TSD. He concluded that this differentiation in climate causes divergent selection on regulatory elements in the sex-determining network allowing for the emergence of sex chromosomes in the highlands.

"Temperature sex determination could allow the mother to determine the sex of her offspring by varying the temperature of the nest in which her eggs are incubated. However there is no evidence thus far that sex ratio is manipulated by parental care"