Tuesday, 14 October 2014

Non-genetic inheritance: Scientific debate spanning generations

Originally published in The Australia Times Science Magazine, Vol.2 No.10 (October) 2014

DNA has long been regarded as the heritable material that encodes the instructions for the development of all known organisms. There is no doubt of its importance to life, and its role in evolution is equally fundamental.

Since the seminal works of researchers including Charles Darwin and Gregor Mendel, it is a widely held belief that changes in gene expression that develop through life experience can not be passed on the next generation. That, barring mutation, there is an unaltered continuity of the germ line.

But is DNA the whole story when it comes to inheritance? The study of non-genetic inheritance has a long history of controversy. From early 20th century scientific scandals to very recent findings, the story of nature vs nurture has fascinated many generations of scientists.

And it seems we have again arrived at a major turning point in our understanding.

Midwife Toads and Indian Ink

Perhaps the most polarizing figure in this debate for the last century was the brilliant and equally unorthodox Viennese zoologist Paul Kammerer. Renown for his expertise in reptile breeding and behavior, Dr Kammerer’s controversial research raised some important questions regarding the nature and causes of variation at a time when genetics and modern evolutionary theory were in their infancy and the subjects of intense debate.

The controversial Austrian zoologist Paul Kammerer (1880-1926). Source: Wikipedia    

The tireless work of Dr Kammerer generated several highly contentious findings that appeared to demonstrate that acquired characteristics could be stably transmitted to subsequent generations. Environmentally induced changes in reproductive characteristics and colouration of salamanders, as well as the development of long, droopy siphons by sea squirts following repeated amputations were among his notable claims for the inheritance of acquired traits.

However his most notorious attempt to demonstrate acquired trait heritability is also one of the greatest and most tragic scientific scandals of the last century. The controversy centers on experiments with the midwife toad, Alytes obstetricians conducted between 1894 and 1914.

The male midwife toad, Alytes obstetricians. He carries the fertilised eggs until they are ready to hatch.
Laurent Lebois/Flickr

This grotesque amphibian derives its name from the curious post-copulation male practice of carrying the fertilized eggs on his hind legs until they hatch. The sexual behaviors of toads and frogs often involves the male grasping the female with his front legs in a process known as amplexus, sometimes for weeks until she spawns.

During this process, mature males of certain species develop darkened and callused swellings known as nuptial pads that aid the male to grip the slippery female body. The midwife toad however naturally mates on land, and the comparatively dryer skin of the female means the males don’t require or possess nuptial pads.

The nuptial pad (indicated by the red arrow) on Pelophylax esculentus. Christian Fischer. Wikimedia Commons.    

Kammerer contended that breeding the midwife toad in water for several generations induced the formation of nuptial pads similar to naturally aquatic-mating species. Unlike those spawned on land, eggs produced in water did not attach to the male’s legs and floated away. From the surviving eggs, Kammerer claimed to have propagated a line of toads that habitually bred in water. Moreover, descendants of these toads were reported to develop nuptial pads as hereditary features that became more prevalent with each generation.

Immense interest saw his findings sensationalized by the press. However Kammerer’s work was met with a high level of academic scrutiny and his attempts to convince the scientific community were not aided by the intellectual climate before the First World War. Paralleling the political and philosophical revolutions of the region in the late 19th and early 20th centuries, the milieu of biological and particularly evolutionary ideas in Europe was at once passionate, severe, fanatical, zealous, vehement, and sometimes sinister. 

Emerging concepts were heavily influenced by the work of many prominent scientists including that of Charles Darwin, the rediscovery of Gregor Mendel’s genetic principles in 1900, as well as the discovery of sex chromosomes in 1905. It was this period of important scientific discoveries and fierce debate that fostered our current understanding of heredity.

Famously proposed by the French naturalist Jean-Baptiste Lamarck in the early 1800’s, the idea that useful adaptations arising in an organism can be stably transmitted to their offspring had been largely usurped by the rise of neo-Darwinism, a concept that depends on the chance occurrence of random DNA mutations. Particularly August Weismann’s theories exemplified by crude experiments that demonstrated a separation of the germ-line from environmental influences (still called the Weismann barrier today) were perceived as a telling blow for Lamarckism. However debates surrounding the formation of modern genetics had fostered the re-emergence of arguments in favour of acquired heritability in the late 1800’s.

The German evolutionary biologist August Weismann (1834-1914).
Source: Wikipedia

Whether Kammerer was a strict Lamarckian is open to debate, with some sources suggesting that his concepts of evolution, though unorthodox, were pursued in a Darwinian framework. In fact many versions of Darwinism and alternative models of evolution were being contested at this time. Despite this, he was billed as a champion of Lamarckism in Stalin’s Soviet Union, later becoming the subject of the heavily distorted socialist realist film Salamandra (1928).

Resistance to Kammerer’s ideas arrived from many prominent biologist in Europe as well as America, by whom he was labeled such things as neo-Lamarckian heretic and pseudoscientist. Among several noteworthy skeptics, the pioneering British geneticist William Bateson was perhaps his fiercest detractor. The effect on Kammerer was severe, driving him to suicide on 23rd September 1926 at the age of forty-five.

Several separate accounts provide different insights into the circumstances surrounding the abrupt end to the career of Paul Kammerer, often distinguished by their emphasis on political and social influences on the scientific rejection of his work. However they all agree on one crucial point - allegations that he artificially tampered with his specimens proved to be Kammerer’s ultimate undoing.

Upon close examination American herpetologist G.K. Noble concluded that, in an apparent attempt to preserve and emphasize the heritable characteristics of the single remaining specimen, Indian ink had been injected into the nuptial pads. Allegations of fraud were vociferous. Fingers were pointed in accordance with the usual practice of scapegoating research assistants. And much to the delight of Dr Kammerer’s academic opponents, Noble published his detailed assessment and condemnation of this deception in the prestigious scientific journal Nature in August 1926.

Others have defended Kammerer’s name. Most notably, Arthur Koestler recounts this story in The Case of the Midwife Toad (1971), where he cites supporting evidence from the personal testimonies of biologists who saw the infamous nuptial pads, as well as photographs of the specimens. Furthermore, Koestler alludes towards a conspiracy to defame the Austrian biologist.

Undoubtedly Paul Kammerer was a peculiar scientist known to entertain remarkably absurd notions of acquired characteristics in humans, including parental interest in music manifesting as musical talent in children, and promising that enforcement of American prohibition would eliminate the next generation’s desire to consume alcohol. His experiments have never been successfully reproduced, though his supporters cite his unparalleled expertise in maintaining live animals as the source of this discrepancy.

Amid further allegations of image manipulation as well as scientific evidence apparently contradicting his findings in sea squirts, Kammerer’s contributions have been largely relegated from scientific discussions. It is difficult to dispute the volume of evidence against his controversial findings. It is equally difficult to dispute that many features of the contemporary academic and social environment were particularly resistant to Kammerer’s radical notions.

Contrasting the unalterable hereditary material that formed the basis of neo-Darwinism, a major barrier to the concept of heritable acquired traits in this formative period was the lack of a clear mechanism. It would be several decades before Conrad Waddington would employ the terms epigenetics and genetic assimilation to crudely describe processes underlying cellular differentiation from a single genome and the evolution of environmentally accommodating phenotypic plasticity.

On the back of many important breakthroughs, these terms have more recently been employed in descriptions of heritable changes in gene expression that arise separately from changes to the DNA code. Experiments have revealed numerous molecular mechanisms associated with epigenetic regulation that could provide the missing link for the transmission of acquired traits.

At least conceptually, Kammerer may have been on to something.

An Epigenetic Mechanism

In his important publication What genes can’t do (2003), Prof Lenny Moss described how biological science “gyrates to a centennial beat.” The term biology was first applied at the beginning of the 19th century. Modern genetics was born early in the 20th century. The start of our current century heard the first descriptions of the sequenced human genome. Now, one hundred years after the important period of scientific debate around the time of the First World War, seminal observations of gene-environment interactions and a pronounced shift to systems level perspectives of biology have fundamentally changed the way scientists view heredity and the function of genetic material.

We now know that the language of the DNA code is not the entire story when it comes to cellular differentiation and the establishment of distinct gene expression profiles. In the cell nucleus, DNA is packaged with small proteins called histones that regulate its compaction. Together, DNA and histones form a dynamic polymer called chromatin. And it is the degree of chromatin accessibility that determines if a gene (a defined stretch of DNA) can be read in order to instruct the production of the protein or functional RNA molecule that it codes for.

A continually expanding plethora of chemical modifications occurring on different components of the chromatin collaboratively underlies this additional level of genomic control. Major players in this intricate orchestra have been identified in a range of animal phyla from worms and flies to rodents and primates, including humans.

Addition of small methyl groups (composed of one carbon and three hydrogen atoms) directly to the DNA generally corresponds with closed chromatin structures and gene silencing, and represents the most accurate etymological interpretation of epigenetics. These methyl groups, as well as various other chemical modifications are dynamically written to and erased from specific sites on the histone proteins by an army of enzymes and protein complexes to also regulate chromatin architecture.

Textbook descriptions of gene expression traditionally assert the classical biological dogma: DNA codes for RNA (transcription), which codes for protein (translation). This basic tenet still holds true at the simplest level. However recent years have revealed that while only a fraction of the genome (approximately 1.5%) codes for protein, more than 90% of the genome is transcribed to RNA. So what are all these non-coding RNA molecules doing?

Research has demonstrated that they can participate at many levels to confer an added layer of complexity to gene expression. The varying physical lengths of different classes of non-coding RNA permit distinct regulatory functions. Short transcripts are associated with controlling the cellular levels of coding RNA molecules, and therefore protein expression. And some of the longer non-coding RNAs have recently been implicated in directing certain epigenetic modifiers to specific sites on the chromatin.

These major components regulate gene activity largely independently of the DNA code. Crucially, epigenetic modifications respond and adapt to environmental stimuli. Furthermore, chromatin modifications can be stably transmitted through cell division and are thus described as heritable. Does this mean that epigenetics could be the vehicle of non-genetic inheritance? The existence of epigenetic inheritance is well established in plants. However animals are an entirely different story.

Renewed speculation of this topic is often described as resurrecting the ghosts of Lamarckism. Similarly reigniting strong debate, Alexander Vargas asked “Did Paul Kammerer discover epigenetic inheritance?” in an article published in the Journal of Experimental Zoology in 2009. We are a long way from answering this specific question. Predominantly because of time scale on which Kammerer’s experiments were conducted. On the other hand, independent commentaries published in that same journal strongly criticize Vargas’ interpretation.

Other experimental animals may provide more manageable models to examine these phenomena. In fact recent studies have turned up potentially revolutionary findings.

Does This Coat Make Me Look Fat?

An allele is one of a number of alternative versions of the same gene. Individuals inherit two alleles for each gene, one from each parent. There is generally a clear dominance of some alleles over others that determine which version of each gene is used. Important experiments had revealed that the expression of some alleles was actually quite variable, not only within an individual but also across genetically identical populations. It turns out that epigenetic modifications control these epi-alleles.

One model in particular has been very useful to study epialleles. Expression of the agouti viable yellow allele (Avy) of the agouti gene in mice results in a yellow coat colour. Buried within this allele is a regulatory stretch of DNA that can be epigenetically methylated. This silences the Avy allele, causing the expression of a brown coat colour.

Whether this allele is expressed or silenced is probabilistic. The proportion of each coat colour varies among Avy mouse littermates ranging from brown to mottled to yellow, despite their identical genetic backgrounds. Furthermore, the yellow coat colour is associated with obesity as well as increased susceptibility to diabetes and cancer. When Avy mice were supplemented with dietary factors known to influence epigenetic methylation reactions, a clear shift toward brown coat colour was observed.

These mice are genetically identical. The epigenetic methylation status of the Avy allele of the agouti gene controls coat colour and the susceptibility to metabolic disease.
Source: Wikipedia
While this appears to be an example of epigenetic inheritance, epigenetic changes manifesting in the offspring are not inherited as such, but are in fact shaped by environmental conditions in utero. By contrast, studies show that yellow mothers are more likely to produce yellow offspring. And by transferring embryos between mothers with different coat colours, researchers found that this effect was independent of the uterine environment. 

A Familiar Fear

A much stronger case for epigenetic inheritance was beginning to build. And late in 2013 some extraordinary findings were reported.

Researchers at the prestigious Emory University in Atlanta managed to condition male mice to fear a specific odour called acetophenone by pairing it with mild electric shocks to the foot. The mice were then allowed to mate and the response to acetophenone was measured in future generations.

Remarkably, the offspring (F1 generation) of these mice showed the same heightened response to the odour despite having never encountered it before. Even more surprisingly, this fear response persisted in to the next generation of male mice (the F2 generation). That these findings were in fact inherited and not a result of social factors was validated by repeating the experiment using in vitro fertilization, which separated the fathers from their offspring.    

A shocking odour. Exposure to acetophenone and foot shock conditions mice to fear the odour. Adult F1 mice (first generation) inherit the fear of acetophenone. The offspring of these mice, the F2 generation, also inherit the conditioned fear response. Adapted from Szyf, M. (2014) Nature Neuroscience. Contains image by George Shuklin/Wikimedia Commons. Image by S. Keating

The researchers were able to narrow down this response to a pathway controlled by the Olfr151 odourant receptor. When they analyzed the sperm of the F0 mice originally condition to fear acetophenone, they found significantly lower DNA methylation at the gene encoding Olfr151. Importantly, this altered methylation pattern was also found in the sperm from the F1 mice.

In the same year a separate study reported that widespread DNA methylation changes in sperm was associated with the transmission of metabolic dysfunction from diet-induced obese mice to two subsequent generations. This study was also interesting for another reason. Remember the non-coding RNAs? Well it turns out that the testes of obese males in the founder generation had significantly altered the levels of hundreds of short non-coding RNAs called microRNAs.

Because they are present in sperm, the idea that non-coding RNA could participate in transgenerational inheritance is rapidly gaining momentum. Furthermore, these molecules could partly explain the inheritance of DNA methylation states that appears to occur despite the well-described erasure of epigenetic modifications that occurs during zygote development. A specific group of non-coding RNAs called piRNAs are thought to guide DNA methylation patterns in mouse sperm.

The Next Generation of Discovery

Transgenerational epigenetic inheritance is real. However we do not yet understand the specifics of this system in the context of different stimuli or in different organisms. The immense interest surrounding transgenerational epigenetics is likely to turn up some very interesting results in the near future, as well as some more polarizing characters.

Exciting research coming out of the laboratory of Michael Skinner at Washington State University is greatly expanding our understanding of the transgenerationally persistent effects of pesticides and other chemicals. Not everyone is convinced though and his research has ruffled more than a few feathers. Data fabrication by one of his staff leading to retraction of an article in 2006 continues the controversial tradition of this contentious area of research.

Going forward there are some important points that should be considered in discussions of epigenetic inheritance.

In light of recent discoveries, many people have been quick to herald the return of Lamarckian evolution. A key consideration here is the timescale required for evolutionary change. The research discussed here has been conducted over a relatively short term. In fact for many mammalian studies of transgenerational inheritance, the epigenetic changes have not been shown beyond the 3rd or 4th generations.

Acquired traits are supposed to be permanent. And most studies do not have the scope to genuinely address this issue. At the moment there is no way to rule out if such epigenetic changes might be reversed, spontaneously lost, or even exacerbated beyond this timespan.

The importance of these discoveries for human health is increasingly debated. It is important to consider not only the gross morphological differences between humans and experimental animals, but also the difficulty of controlling for genetic effects in outbred human populations.

Finally, the interactions of dietary factors, metabolic pathways, and epigenetics are a hot topic in biology at the moment. Many researchers have speculated on the inheritance of metabolically induced epigenetic changes that may predispose offspring to disease in adulthood. While this is a very important topic in itself, these studies are usually not examples of epigenetic inheritance, but can be more accurately described as epigenetic gestational programming.

Debates about non-genetic inheritance have persisted for more than 200 years. The next few years may be the most important as our understanding of epigenetics develops.


Koestler, A (1971) The Case of the Midwife Toad. Random House.

Logan, C. A. (2013) Hormones, Heredity and Race: Spectacular Failure in Interwar Vienna. Rutgers Press.

Noble, G. K. (1926) Kammerer’s Alytes. Nature 118: 209-211

Kammerer, P. (1926) Paul Kammerer’s letter to the Moscow Academy of Sciences. Science 64:493-494

Weissman, G. (2010) The Midwife Toad and Alma Mahler: Epigenetics or a Matter of Deception? The FASEB Journal 24(8): 2591-5

Moss, L. (2003) What genes can’t do. The MIT Press.

Hauser, M.T., Ausfatz W., Jonak C., Luschnig C. (2011) Transgenerational epigenetic inheritance in plants. Biochimica et Biophysica Acta 1809(8): 459-68

Vargas, A.O. (2009) Did Paul Kammerer discover epigenetic inheritance? A modern look at the controversial midwife toad experiments. Journal of Experimental Zoology 312(7): 667-78

Gliboff, S. (2010) Did Paul Kammerer discover epigenetic inheritance? No and why not. Journal of Experimental Zoology 314(8): 616-24

Svardal, H. (2010) Can epigenetics solve the case of the midwife toad? – A comment on Vargas. Journal of Experimental Zoology 314(8): 625-8

Mercer, T.R., Mattick, J.S. (2013) Structure and function of long noncoding RNAs in epigenetic regulation. Nature Structural and Molecular Biology 20(3): 300-7

Wolf ,G.L., Kodell, R.L., Moore, S.R., Cooney, C.A. (1998) Maternal epigenetics and methyl supplements affect agouti gene expression in Avy/a mice. The FASEB Journal 12(11): 949-57

Dias, B.G., Ressler, K.J. (2014) Parental olfactory experience influences behavior and neural structure in subsequent generations. Nature Neuroscience 17(1): 89-96

Szyf, M. (2014) Lamarck revisted: epigenetic inheritance of ancestral odor fear conditioning. Nature Neuroscience 17(1): 2-4

Fullston, T., Ohlsson Teague, E.M., Palmer, N.O., DeBlasio, M.J., Mitchell, M., Corbett, M., Print, C.G., Owens, J.A., Lane, M. (2013). Paternal obesity initiates metabolic disturbances in two generations of mice with incomplete penetrance to the F2 generation and alters the transcriptional profile of testis and sperm microRNA content. The FASEB Journal 27(10): 4226-43

Kaiser, J. (2014) The epigenetics heretic. Science 343(6169): 361-3


Image 1: http://en.wikipedia.org/wiki/Paul_Kammerer#mediaviewer/File:Kammerer.jpg

Image 2: Laurent Lebois, “Alytes obstetricans almogavarii (Pyrénées Orientales 66)”, Flickr, copyright 2007 under an attribution licence, https://www.flickr.com/photos/cheloran/4011058121

Image 3: Christian Fischer, “A male Edible Frog”. Wikimedia Commons, copyright 2010 under an attribution licence, http://commons.wikimedia.org/wiki/File:NuptialPad.jpg

Image 4: http://en.wikipedia.org/wiki/August_Weismann#mediaviewer/File:August_Weismann.jpg

Image 5: “Agouti mice” by Randy Jirtle and Dana Dolinoy. http://en.wikipedia.org/wiki/Randy_Jirtle#mediaviewer/File:Agouti_Mice.jpg

Image 6: Incorporates image by George Shuklin. “Mus musculus”. Wikimedia Commons, copyright 2008 under an attribution licence, http://en.wikipedia.org/wiki/File:Мышь_2.jpg