Platypus: A Darwinian Cautionary Tale

The platypus is an evolutionary enigma. According to Professor Tom Kemp: The greatest mystery of all concerning mammalian evolution stretches back for 200 years: the question of what exactly the monotreme mammals are, and how they relate phylogenetically to therians [from T.S. Kemp (2007) The Origin and Evolution of Mammals Oxford University Press pages 173-174].

Mammals comprise three major groups. The therians [theria] include all placental mammals [eutheria] and marsupials [metatheria] which have the characteristic pouch in which the immature foetus is nurtured after birth. The third group of mammals are the monotremes or egg-laying mammals. These are the prototheria [as distinct from the theria] and there are only three species of monotremes living today, only found in Australia. These are the platypus [Ornithorhyncus], the short beaked echidna [Tachyglossus] and the long-beaked echidna [Zaglossus]. 

Is it a reptile?

At one time, it was believed that the monotremes were transitional between reptiles and mammals. It is easy to see why. The prototheria [Gk: first beast] lay small round eggs and have some other reptilian features. For example, the male platypus possesses a spur on its hind limbs through which it delivers a venomous cocktail produced by the crural glands located in its upper thigh. This venom contains hundreds of different chemicals including four major toxins. Three of these toxins are unique to the platypus and are described as defensin-like proteins [DLPs]. 

Whittington et al. have recently published a paper in the academic journal Genome in which they describe significant homology [structural similarity] between the DLPs and proteins present in the venom of snakes and other reptiles. However because of the prevailing evolutionary phylogenetic classification that places reptiles and monotremes in very different clades, the general consensus is that similar venom proteins evolved independently in reptiles and monotremes. This is considered an example of convergence which suggests that given the appropriate conditions, similar proteins and structures will evolve any number of times in different organisms. In their own words: Convergent evolution has repeatedly selected genes coding for proteins containing specific structural motifs as templates for venom molecules.

With respect to Whittington and his colleagues, at least they attempt to speculate how this might have occurred by suggesting processes involving "gene duplication and subsequent functional diversification". Presumably, functional diversification must involve the repeated emergence of active protein intermediates with some selective advantage. It is interesting to ponder what these unknown proteins might have been and what particular function they might have served. Unfortunately, this is rarely done. 

Nevertheless, the main reason that monotremes cannot be considered intermediate between reptiles and mammals is the current Darwinian consensus that reptiles and mammals have evolved independently from a putative common ancestral amniote either via the synapsid [to leading to mammals] or sauropsid [leading to reptiles and birds] lineages. We refer you to the article "Synapsids and the Evolution of Mammals" which can be found here.

Consequently, it is something of an embarrassment to describe the synapsids as "mammal-like reptiles" which is still a very common occurrence in the both the academic and popular press. As mentioned in the previous article, Professor Donald Prothero is dogmatic about this anomaly. In chapter 13 entitled "Mammalian Explosion" in his book, he writes as follows: Of the transitional series that we have examined between major groups of vertebrates, one of the best documented is the transition from primitive amniotes to mammals via the synapsids, formerly known as the "mammal-like reptiles." As we explained previously, however, the synapsids that evolve into mammals are not reptiles and never had anything to do with the lineage that leads to reptiles ...This idea is now completely discredited, and anyone who still uses the obsolete and misleading term mammal-like reptiles clearly doesn't know much about the current understanding of vertebrate evolution [from Donald R. Prothero [2007] Evolution: What the Fossils Say and Why It Matters New York: Columbia University Press page 271].Tom Kemp goes on to consider the mystery of the monotreme evolution: The relationship of monotremes to the Mesozoic mammal groups is considerably less clear, and the development of views about this problem has had an extraordinarily chequered history. At one time it was believed by almost everyone that monotremes had a separate origin from pre-mammalian therapsids, implying convergent evolution of their mammalian characters. The discovery of late Triassic mammals of South Wales quickly altered that view because similarities were seen between, on the one hand Morganucodon and the monotremes and on the other Kuehneotherium and the living therians [from T.S. Kemp (2007) The Origin and Evolution of MammalsOxford University Press page 175].As we reported in our previous article, Morganucodon is generally regarded as a mammal although it also possesses some reptilian features, particularly in its lower jaw. The reason why this creature is considered a relative to the monotremes is its teeth. As Kemp points out, this is highly contentious because the platypus sheds its juvenile teeth which are replaced in maturity by bony ridges. To complicate matters further, echidnas do not have teeth at all. The other ancient mammal referred to by Kemp is Kuehneotherium which has been found in exactly the same strata in South Wales [for further information see Kemp pages 162-163]. After a consideration of academic attempts to relate monotremes to Morganucodon, Kemp concludes: Compared to this very weak evidence for monotreme-Morganucodon relationship, there are several characters shared by monotremes and therians that can be demonstrated to be more derived than in Morganucodon ... All cladistic analyses now place monotremes closer to living therians than to morganucondontids [from T.S. Kemp (2007) The Origin and Evolution of Mammals Oxford University Press page 176].Since the recent publication [May 2008] of the platypus genome in the journal Nature there has been a great deal of activity in the academic press and in the popular media including the BBC. All this activity, however, has not brought any resolution to the evolutionary enigma that is the platypus. If anything, the situation has become even more confused. For example, Elizabeth Finkel wrote a review entitled "Genome speaks to Transitional Nature of Monotremes" in the prestigious American journal Science in which she states: The clearest traces of the journey from reptile to mammal come from tracking the yolk and milk genes. Chickens have three vitellogenin egg yolk genes; the platypus has just one left. But the casein milk protein genes that mammals have but reptiles don't are all there. And just as in other mammals, in platypus, they are clustered next to the tooth enamel genes from which they are thought to have evolved, the researchers report.Here in this statement, one can discern something of the current confusion in attempts to trace the evolutionary history of the monotremes. The suggestion here is that the platypus has evolved from a reptilian lineage which is anathema according to Professor Prothero. Current Darwinian thinking dictates that mammals have evolved from synapsids which are not reptiles. Of course, presumably, synapsids were oviparous and like all egg-laying animals would have possessed yolk proteins and the vitellogenin coding genes. This is a reasonable assumption, since according to Patrick Brabin in his recent paper in Gene [May 2008], vitellogenin homologues have been found in amphibians, fish, birds as well as in the platypus and reptiles. 

On the other hand, therian mammals [i.e. both placental and marsupial] do not lay eggs and it might be thought that they have no need of vitellogenin genes. 

In particular, David Brawand et al. have published a recent paper [March 18 2008] in the open access journal PLoS Biology entitled "Loss of Egg Yolk Genes in Mammals and the Origin of Lactation and Placentation". In the abstract of this paper they write: Embryonic development in nonmammalian vertebrates depends entirely on nutritional reserves that are predominantly derived from vitellogenin proteins and stored in egg yolk. Mammals have evolved new resources, such as lactation and placentation, to nourish their developing and early offspring. However, the evolutionary timing and molecular events associated with this major phenotypic transition are not known. By means of sensitive comparative genomics analyses and evolutionary simulations, we here show that the three ancestral vitellogenin-encoding genes were progressively lost during mammalian evolution (until around 30–70 million years ago, Mya) in all but the egg-laying monotremes, which have retained a functional vitellogenin gene.

Their research identified vitellogenin pseudogenes in the previously published human and dog genome. A pseudogene is essentially a non-active version of a previously functional gene that has been switched off. In the case of the vitellogenin genes, Brawand and his group found that there were "premature stop codons and frame-shifting insertion/deletions" in genomic regions equivalent to the active vitellogenin genes found in the chicken. Are these pseudogenes a relic of mammalian evolutionary history? Brawand and his colleagues think so but is their another explanation? 

Do placental and marsupial mammals need the ability to produce egg yolk at any stage in their life cycles. The answer is an emphatic yes! In the paradoxical words of Brawand and his group: Marsupials also have a placenta, originating from the yolk sac, but the marsupial oocyte [egg] contains considerably more yolk than that of eutherians, which is virtually devoid of it. The marsupial yolk reserve is assumed to be essential during the earliest development of the embryo, complementing the uptake of uterine secretions by the yolk sac, prior to shell coat rupture. However, the content of marsupial yolk is not well known.The fact is this; all mammals require yolk at some stage in development. In particular, all placental mammals produce eggs that require yolk for nourishment prior to the establishment of the placenta. In the words of LM Baggott in his book entitled Human Reproduction: The eggs of mammals contain relatively little yolk compared to the eggs of other vertebrates. However, yolk is present in sufficient quantity to sustain the development of the embryo through the period of cleavage ... In placental mammals, including of course humans; cleavage takes place as the embryo passes down the Fallopian tube towards the uterus. During this period, the embryo draws upon the reserves of yolk in the dividing cells. After implantation, has taken place in the uterus and until birth, energy and raw materials come to the developing embryo from the maternal circulation [from LM Baggott (1997) Human Reproduction Cambridge University Press page 33].

This fact might allow an alternative explanation for the presence of vitellogenin pseuodogenes in placental mammals. In fact, seen in this light, pseudogenes may sometimes be active genes that are permanently switched off in many different types of cell during development and cellular differentiation. For example, a skin cell has no need of yolk but will continue to possess the full genetic compliment. In this particular cell, however, the vitellogenin gene will have been inactivated. What is much more difficult to consider from a Darwinian perspective, however, is the emergence of a fully functional placenta with everything associated with it. It is almost impossible to comprehend what would have to take place to move from an oviparous (egg-laying) to a viviparous (live birth) physiology. Brawand et al. do not begin to consider this even though they include the word "placentation" in the title and abstract of their paper. 

On the other hand, like all mammals, the platypus possesses the metabolic and physiological capability to produce milk. Brawand and colleagues appear somewhat bemused by the fact that the platypus has casein (milk protein) genes. They state: We screened the platypus genome to see whether monotremes do in fact have orthologous casein genes, which would imply that these genes emerged in the common mammalian ancestor. Interestingly, we identified three putative casein genes in a genomic region that is syntenic to that carrying the casein genes in therians.Two genes are said to be orthologous [a Darwinian term] if they diverged after a speciation event. A syntenic genomic region is the location on different chromosomes from different species possessing similar genes. The language of the Brawand paper is remarkable "we identified three putative casein genes in a genomic region that is syntenic to that carrying the casein genes in therians". It is as though they were surprised to find them. The fact is all mammalian lactation requires that all the genes for the production of all the protein constituents of milk are present and that the physical structures necessary to deliver that milk are functional. The platypus, however, does not suckle its young in the conventional mammalian manner. It does not have nipples but exudes milk from specialised glands on its abdomen. These glands are generally regarded as modified sweat glands. For example, Lewis Wolpert in his article in The Independent [December 8 2004] wrote: One important piece of evidence is the platypus, a monotreme mammal that has a patch on its breast that secretes milk for its infants to suck ... Clearly, the platypus was an early stage in breast evolution. Numerous theories have been proposed for how lactation evolved in the platypus. One theory, more than a hundred years old, suggests that the glands secreting nourishment in the platypus are modified sweat glands, but that the lactating glands in other mammals are modified sebaceous glands, which normally secrete an oily fluid to protect the skin. In the 1960s, JBS Haldane took up the problem and proposed that the ancestors of monotreme mammals might have needed to keep their eggs cool, and so evolved a mechanism for moistening them in their fur, in a manner similar to that used by some Asian birds who moisten their feathers. More recently a theory has emerged in which multiple glands of the skin are involved.

According to Professor Wolpert and others, lactation in the platypus was an "early stage in breast evolution" perhaps being required to keep eggs cool by secreting milk from modified sweat glands. It is worth evaluating this hypothesis objectively.

Sweat glands are characteristic of warm blooded mammals. Furthermore, reptiles [and birds] do not possess sweat glands. On the other hand, mammalian skin has two types of sweat glands: apocrine and merocrine. The secretion from both these types is controlled by the endocrine and the autonomic nervous systems. Apocrine sweat glands produce a thick secretion containing pheromones. Merocrine sweat glands are more widely distributed and are closely involved in temperature regulation and excretion by secreting sweat which is 99 percent water.

So we have a few questions: 

1. For the lactation machinery to evolve from a sweat gland there has to be a functioning sweat gland but reptiles [and birds] do not possess sweat glands. There is no requirement for sweat glands until the animal has evolved an endothermic physiology. In other words, a mammal has to be a mammal to possess a sweat gland. 

2. Since sweat is 99% water, why was it deemed necessary to evolve all the milk protein coding genes just to keep the eggs cool? It is worth noting that milk has no nutritional benefit for a developing embryo encased in the monotreme egg.

If, as Professor Wolpert and others suggest, lactation in the platypus is the beginning of the evolution of the breast [presumably the nipple and the machinery associated with it], one has to assume that all the ancient mammals from the mid-Triassic to the Cretaceous had monotreme-like oviparous physiology even though there is no fossil evidence to suggest that this is true. The general Darwinian consensus [also reflected in the Nature paper on the platypus genome] is that the monotremes diverged from the other mammals approximately 166 million years ago.

Nevertheless, the oldest known fossils recognisable as monotremes include Teinolophos trusleriSteropodon galmani and Kollikodon ritchiei. These creatures are conventionally dated at approximately 100 to120 million years, supposedly 100 million years after the emergence of true mammals in the fossil record in the late Triassic.

The situation is made even more confusing by the discovery of several mammalian fossils in the southern hemisphere that possess tribosphenic [three-cusped] molars. These teeth are characteristic of all marsupials and placental mammals which were thought to have emerged initially only in the northern hemisphere. For example, Thomas Rich of the Museum of Victoria in Melbourne spent many years looking for the ancestors of Australia's mammals. In the late 1990s he discovered an ancient jaw which he and his co-workers subsequently described in a publication in the academic journal Science. In the abstract we read: A small, well-preserved dentary of a tribosphenic mammal with the most posterior premolar and all three molars in place has been found in Aptian (Early Cretaceous) rocks of south-eastern Australia. In most respects, dental and mandibular anatomy of the specimen is similar to that of primitive placental mammals.

Rich called this animal Ausktribosphenos nyktos and since that time several other fossils have been found in Australia (Bishops), Madagascar (Ambondro mahabo) and South America (Asfaltomylos) that seem to indicate an emergence of placental animals in the fossil record that predates the most ancient monotremes. This has been extremely controversial. For example, Luo et al., proposed independent convergent evolution for tribosphenic teeth in the northern and southern hemispheres. This was reported in Nature

The most extensive cladistic analysis of mammalian fossil dental characteristics has been undertaken was published in 2003 in the journal Molecular Phylogenetics and Evolution. In the words of Kemp: Woodburne et al. (2003) undertook a cladistic analysis, based on 51 characters, mostly dental but a few mandibular. They found that monotremes, including Steropodon and Teinolophos as basal members, are a sister group of all the therian mammals. Furthermore the disputed genera AmbondroAusktribophenosAsfaltomylos, and Bishops constituted a monophyletic group that nests within the stem placentals [from T.S. Kemp (2007) The Origin and Evolution of Mammals Oxford University Press page 180].

More recently, another "mammal-like" fossil has been discovered in mid-Jurassic sediments in Mongolia which may eventually lead to a total revision of our understanding regarding the emergence of mammals. The creature has been called Castorocauda lutrasimilis

This discovery was first described by Ji et al. in a 2006 publication entitled "A Swimming Mammaliaform from the Middle Jurassic and Ecomorphological Diversification of Early Mammals" in the journal Science. They conclude their descriptive paper as follows: Castorocauda was a semiaquatic carnivore, similar to the modern river otter. This fossil shows that basal mammals occupied more diverse niches than just those of small insectivorous or omnivorous mammals with generalized terrestrial locomotory features. Castorocauda also suggests that mammaliaforms developed physiological adaptations associated with pelage [fur], well before the rise of modern Mammalia, and had more diverse ecomorphological adaptations than previously thought, with at least some lineages occupying semiaquatic niches.At the present time, Castorocauda is not regarded as a true mammal. According to current Darwinian thinking, the creature can only be regarded as a mammaliaform or proto-mammal. Nevertheless, the most remarkable feature of this fossil discovery is the preservation of its fur. According to Ji et al.: The fur of Castorocauda is preserved as impressions of guard hairs and carbonized under-furs. Hairs and hair-related integument structures are important characteristics of all modern mammals ... the broad and scaly tail of Castorocauda was similar to that of the modern beaver Castor canadensis, a semiaquatic placental mammal well adapted for swimming.The presence of fur is a clear indication that the animal was warm-blooded and although the animal appears most like a semi-aquatic placental mammal (e.g. beaver or otter), the authors suggest that the forelimbs are similar to the platypus in that they are adapted for both digging and swimming. There is also the indication of webbing on the hind feet. Furthermore, the National Geographic once had an article that stated: Even tiny middle-ear bones are intact. The well-preserved teeth - incisors, canines, premolars, and molars - look to have been ideal for feeding on fish and aquatic invertebrates, somewhat like the teeth of modern seals ... Castorocauda has the ankle spurs characteristic of its nearest living relative, the platypus, which uses them for territorial defense. And like the platypus, Castorocauda was probably an egg-layer.

So was Castorocauda an ancient monotreme? According to Kemp, "the solution to the mystery of the monotremes continues to be elusive". Presumably, this will remain the situation until there is some major revision in current Darwinian thinking.

Is it a bird?

Perhaps the most remarkable feature of the platypus genome is the structure and number of sex chromosomes. Typically, all male mammals have one X and one Y [i.e. heterogametic] chromosome whereas females possess two X [i.e. homogametic] chromosomes. The male platypus has five X and five Y chromosomes and the female platypus five pairs of X chromosomes. In most mammals, the Y chromosome possesses a gene called SRY which is a major sex determining factor but this appears to be absent in the platypus. In addition, there also appears to be no homology between the X chromosomes of placental mammals and the platypus. Sex determination in the platypus is therefore something of a mystery and a great deal of research is continuing in order to discover its mechanism.

Several groups have reported the identification of a gene called DMRTI on the X5 chromosome of the platypus. DMRTI is thought to be a sex determining factor in birds! Avian sex chromosomes comprise Z and W [as distinct from X and Y]. Unlike mammals, male birds carry paired homogametic sex chromosomes [ZZ], females have unpaired heterogametic [ZW] chromosomes. DMRTI is found on the Z chromosome of birds. The double dose of DMRTI in male birds is thought to be a trigger for sex determination. Elizabeth Finkel suggests: The story of the platypus' march away from the reptilian world is also told in the sex chromosomes. According to Jenny Graves of the Australian National University in Canberra, sex chromosome wise, "they do it like a chicken" ... The genome sequence now shows that one of the platypus X chromosomes [X5] has more than just that one bird gene: It's almost entirely equivalent to the chicken Z chromosome. More recent evidence supporting the contention that sex determination in the platypus is similar to that in birds has just been published in Genome Research [June 2008]. In the abstract, Veyruns et al. state: Most significantly, comparative mapping shows that, contrary to earlier reports, there is no homology between the platypus and therian X chromosomes. Orthologs of genes in the conserved region of the human X (including SOX3, the gene from which SRY evolved) all map to platypus chromosome 6, which therefore represents the ancestral autosome from which the therian X and Y pair derived. Rather, the platypus X chromosomes have substantial homology with the bird Z chromosome (including DMRT1) and to segments syntenic with this region in the human genome. Thus, platypus sex chromosomes have strong homology with bird, but not to therian sex chromosomes, implying that the therian X and Y chromosomes (and the SRY gene) evolved from an autosomal pair after the divergence of monotremes only 166 million years ago. Therefore, the therian X and Y are more than 145 million years younger than previously thought.So do the platypus and the other monotremes share any common ancestry with birds? According to the current Darwinian hypothesis, birds are viewed as living dinosaurs, in that they are thought to have descended from sauropsid ancestors. As mentioned in our previous article, the sauropsids are distinct from the synapsids, the supposed ancestors of the mammalian line. However, this model has become even more impenetrable with the recent suggestion that there are multiple independent origins for sex determination in amniotes. The full paper is available here. In the review of this work, Vallender and Lahn suggest: It is generally accepted that environmental sex determination is the ancestral state and that genetic sex determination evolved as a derived condition. It is also recognized that genetic sex determination is evolutionarily highly labile, having evolved into existence on many independent occasions across diverse taxa[emphasis added]. A case in point is sex-determination mechanisms in amniotes (a clade encompassing reptiles, birds, and mammals). The ancestral state in amniotes is likely temperature-dependent sex determination, which is still found in many extant reptilian species, such as crocodilians and some turtles and lizards. From this ancestral state, genetic sex determination evolved in birds, which utilize the ZZ:ZW system, and also independently in mammals, which use the XX:XY system.This problem with particular reference to the platypus has been discussed in detail in the most recent review by Wallis et al. published in the journal Cellular and Molecular Life Sciences [June 26 2008]. They conclude their review as follows: Interest in elucidating the sex-determining system from which SRY assumed control in therians has intensified following our recent finding that the sex chromosomes of birds and monotremes share homology. The possibility that the ancestor of amniotes harboured a sex chromosome system still maintained in monotremes and birds today, while intriguing, still faces several obstacles: the apparent lack of homology between many amniote sex chromosomes, the frequency of subsequent transitions to temperature sex determination in reptiles, and the inferred sex heterogamety transition using the same sex chromosomes. 

In these last two quotes; we can appreciate something of the Darwinian dilemma. Vallender and Lahn suggest that "the ancestral state in amniotes is likely temperature-dependent sex determination" whereas Wallis et al. are tempted to conclude "that the ancestor of amniotes harboured a sex chromosome system still maintained in monotremes and birds today". The sex chromosome system in monotremes [indeed all mammals] and birds does not involve temperature-dependent sex determination. As we have seen, sex determination in reptiles, birds, monotremes and the therian mammals [both marsupial and placental] are all very different and distinctive. Furthermore, with these independent, multiple and, in some cases, convergent evolutionary events supposedly taking place, the maintenance of fertility is of paramount importance. We will consider this vital aspect in some detail in a future article. 

Of course, there are also other genetic features that appear to be unique to the platypus. These include the possession of all the biology required for the exquisitely sensitive chemical and electrical detection systems in its leathery bill. In particular, researchers have discovered numerous genes coding for odour [vomeronasal] receptors. Similar genes are found in many other mammals that rely on a sense of smell, the dog being a classic example. The platypus, however, requires this sensitivity underwater. In 2007, Wendy Grus and her colleagues at the University of Michigan published a major article on these odour receptors in the platypus. In addition, John Pettigrew of the University of Queensland has written a wonderful review entitled "Electroreception in monotremes" in the Journal of Experimental Biology. According to Pettigrew: Its complexity belies the common misconception that monotremes are in some way primitive. The close apposition of mechanoreception and electroreception systems in platypus cortex raises new questions about their relationship.

In other words, the brain of the monotreme is specifically wired to enable the creature to perform its remarkable abilities. It is not at all surprising therefore, following the publication of the platypus genome, that this uniqueness is reflected in the genes of this amazing creature. In fact, whatever the platypus does, it will require the genes to enable it to do so. It is like a bird as it lays eggs. It is like a reptile as it produces venom. It is a warm blooded fur-covered mammal producing milk to suckle its young. It has a unique electro-sensory system and can detect odours and pheromones underwater with unparalleled sensitivity. It is worth noting that the other main group of vertebrates that rely of electroreception are fish and sharks in particular! 

Thus the platypus will remain a significant misfit in any Darwinian scheme. Is it from a sauropsid lineage which includes reptiles and birds? Is it from a synapsid lineage which supposedly led to the emergence of the mammals? Or is it derived independently from some unknown ancestral amniote? Or could it be that the Darwinian hypothesis, cladistic analysis or any other classification system for that matter is just far too restrictive? Without doubt, there are mammal-like reptiles as there are reptile-like mammals. The platypus is a Darwinian cautionary tale. Is it a bird or is it a plain ... old platypus?