Inheritance of acquired characters in animals: A historical overview, further evidence and mechanistic explanations

Abstract

Since the earliest days of evolutionary thought, the problem of the inheritance of acquired characters has been a central debate. Lamarck, Darwin and several of the greatest figures in the history of biology accepted the inheritance of acquired characters as an established fact, but many geneticists refused to accept its reality. Historically speaking, three epochal episodes – Weismann's Barrier, Kammerer's midwife toad and the Lysenko affair – have made this subject a heated scientific and political controversy. Over the past several decades, a substantial body of reliable experimental evidence has accumulated that environmentally induced or acquired changes in animals can be transmitted to future generations. Several well-documented examples, including diet-induced heritable changes, the inheritance of acquired habit, the inheritance of acquired immunity, the inheritance of characters acquired by blood transfusion (also known as vegetative hybridisation), the inheritance of characters acquired by a dam from her former mates (also known as telegony), are briefly reviewed in this article. There are several fundamentally different mechanisms underlying the inheritance of acquired characters and Darwin explained it only with his Pangenesis, a developmental theory of heredity. It can now be understood in terms of molecular genetics such as epigenetic inheritance, prion inheritance, RNA-mediated inheritance and horizontal gene transfer. It is now time to recognise that a new understanding of the inheritance of acquired characters opens a broader perspective on genetics and evolution.

Keywords:

• Inheritance of acquired characters
• Darwin's Pangenesis
• epigenetic inheritance
• RNA-mediated inheritance
• horizontal gene transfer

Introduction

In 1809, the year of Charles Darwin's birth, Jean Baptiste Lamarck published his most important book, La Philosophie Zoologique. It is in this book that Lamarck presented the first scientific evolutionary theory and accepted the hypothesis that acquired characters were heritable. According to Lamarck, ‘all that nature has caused to gain or lose by the circumstances to which their race has been exposed for a long time, and, consequently by the influence of a predominant use or constant disuse of an organ or part, is conserved through generation in the new individuals descending from them, provided that these acquired changes are common to the two sexes or to those which have produced these new individuals’.

Darwin regarded his The Origin as an incomplete explanation of his theory of evolution. Afterwards he supplemented it by his The Variation. It was Darwin's attempt not only to strengthen his theory of natural selection, but more importantly an attempt to describe and understand the root cause of variation. For variation is the fountainhead of evolution. It was the opinion of Darwin that natural selection has been the chief, but not the only, cause of organic evolution. He believed that evolution was due to the natural selection of both innate and acquired characters, and his Pangenesis was an attempt to provide a hypothesis to account for the inheritance of acquired characters more than anything else. Darwin (1868) clearly stated that ‘a multitude of newly-acquired characters, whether injurious or beneficial, whether of the lowest or highest vital importance, are often faithfully transmitted – frequently even when one parent alone possesses some new peculiarity; and we may on the whole conclude that inheritance is the rule, and non-inheritance the anomaly’. Unfortunately, the conventional wisdom is that recognising the inheritance of acquired character was Lamarck's and Darwin's greatest error.

The problem of the inheritance of acquired characters is one that has exercised the minds of biologists since the days of Lamarck and Darwin. Several of the greatest figures in biology and practical breeders, such as Herbert Spencer, George John Romanes, Ernst Haeckel, Ivan Pavlov, Luther Burbank, Ivan Michurin and Trofim Denisovich Lysenko, deeply believed in the inheritance of acquired characters, but many geneticists, such as August Weismann, Hugo de Vries, Wilhelm Johannsen, William Bateson and Thomas Hunt Morgan, strongly opposed it. The decisive successes of Mendelian genetics and molecular biology submerged this debate. Among 30 of the most widely used college textbooks of genetics published during the 1960s–1980s, none indicated that actual examples of the inheritance of acquired characters had been found (Landman 1991). The general plan of this article will be, first, to briefly present a historical overview of the inheritance of acquired characters, and then to review several well-documented examples, and finally to discuss different mechanisms based on Darwin's Pangenesis, epigenetic inheritance, RNA-mediated inheritance and horizontal gene transfer.

Why has the inheritance of acquired characters been so controversial?

Weismann's Barrier

Weismann was one of the most influential thinkers in the history of biology. He explained heredity by his theory of the ‘Continuity of the Germ-plasm’ (Weismann 1912). He suggested that from the very first division of the fertilised egg, one line of cells – the germ-plasm – was distinct from the body cells or somatoplasm. The germ-plasm was unaffected by the somatoplasm or by external influences. A corollary of Weismann's theory is that acquired characters are not inherited. This is referred to as the Weismann's Barrier (Steele et al. 1998). Although the actual continuity of the germ-plasm from generation to generation is clear in certain invertebrate animals, the distinction between germ-plasm and somatoplasm is not clearly evident, and germ cells appear to arise from unspecialised somatic cells in most animals. In addition, Weismann's theory cannot apply to plants because the sex cells of plants may arise from any tissue, and the sex organs are not completely insulated, nor isolated, from the body tissues.

To demonstrate the fallacy of the idea of inheritance of acquired characters, Weismann performed an influential experiment: he cut off the tails of male and female mice, and showed that, in breeding experiments extending over many generations, such tail chopping at birth never produced tailless offspring (Weismann 1904). Critics of this experiment have pointed out that such experiments did not test the inheritance of acquired characters. First, the acquired characters are divided into two kinds: directly acquired characters (such as the removal of a tail) and indirectly acquired characters (such as the modification produced by a change of habit in an organism). Lamarck himself claimed that only characters, which are indirectly acquired, are transmitted (Licorish 1911). Second, Darwin (1868) stated clearly that ‘a part or organ may be removed during several successive generations, and if the operation be not followed by disease, the lost part reappears in the offspring. Dogs and horses formerly had their tails docked during many generations without any inherited effect… . We have as yet spoken only of the removal of parts, when not followed morbid action: but when the operation is thus followed, it is certain that the deficiency is sometimes inherited’. Third, there is good evidence for inheritance of mutilations (Pieraccini 1927; Davenport 1933; Tchang et al. 1964). For example, the occurrence of a genetically stable doublet form has been described in numerous ciliate species. The doublet ciliate phenotype is inherited as a cytoplasmic trait without changes of nuclear genotype, thus providing a remarkable exception to Weismann's germ-plasm theory. Had Weismann chosen to cut Oxytricha doublets in two rather than to amputate the tails of successive generations of mice, surgery in a single generation would have been sufficient to produce an inherited change, as Landman (1993) pointed out.

The case of the midwife toad

It is well known that Paul Kammerer, an Austrian biologist, argued strongly in favour of Lamarckian view on the inheritance of acquired characters (Kammerer 1924). In his most controversial experiment, Kammerer forced midwife toads, which live and mate on land, to mate and lay their eggs in water. Most of the eggs died, but a few of the offspring that survived had lost the terrestrial habits of their parents, and by the third generation they began to develop black nuptial pads on their forelimbs, a character common to water-dwelling species (Pennisi 2009). These experiments generated enormous controversy in the early years of the twentieth century and finally culminated in accusations of scientific fraud (Steele et al. 1998). According to Koestler (1971), these charges were never adequately proven. This episode has been analysed in detail by Gliboff (2005), who believed that Kammerer's evidence was probably genuine. Most recently, Vargas (2009) has re-examined Kammerer's midwife toad experiments and argues that these experiments show signs of epigenetic inheritance.

The Lysenko effect

The third historical influence, which has contributed to making the inheritance of acquired characters a ‘no-go’ zone, can be attributed to the Lysenko effect (Steele et al. 1998). Lysenko was an influential agronomist in the Soviet Union during the Stalin years. He used to regard Mendelian genetics as ‘bourgeois science’ and ‘pseudoscience’ and forced Soviet geneticists to accept Michurin's teaching or be banned from doing research, for which he became notorious (Liu 2004). Lysenko was also a keen supporter of the inheritance of acquired characters. During the 1930s and early 1960s, he led large-scale experiments on graft hybridisation and the conversion of winter and spring wheat, typical examples for the inheritance of acquired characters in plants. Some Western geneticists repeated these experiments and obtained negative results, thus regarding Lysenko as a forger and a criminal (Hagemann 2002). Over the past several decades, however, a considerable body of experimental evidence for Lysenko's claims has been found in scientific literature (Liu 2006; Li & Liu 2010). It is rather an ethical problem whether the scientific works of Lysenko should be ignored (Flegr 2002).

Several examples of the inheritance of acquired characters

Diet-induced heritable changes

It has been shown that dietary and genetic variations during development affect the phenotypes of offspring. Sladden (1938) experimented on the stick-insect Carausius morosus, a native of South India, which in England normally feeds on privet. This insect was subjected to periods of semi-starvation, each lasting two days, when it was offered ivy for food, which it detested. At the end of each period it was revived by a leaf of privet. Each of these ordeals was called a ‘presentation’. The overwhelming majority of the insects would not accept ivy until the fifth presentation, and some even held out until the tenth presentation. These insects are parthenogenetic and when their eggs were collected and bred from, it was found that they accepted ivy at an earlier presentation than did their parents until at the end of five years they took ivy at the first opportunity (Sladden 1938; Macbride 1939).

Changes to the dam's diet during pregnancy can alter the proportion of yellow mice within a litter. When the dam's diet is supplemented with methyl donors (including betaine, methionine and folic acid), there is a shift in the colour of their offspring away from yellow to brown to almost black (Wolff et al. 1998; Wateland & Jirtle 2003). Similar effects have been observed following feeding the dams genistein, which is found in soy milk (Dolinoy et al. 2006). This suggests that the diet of a pregnant female could affect not only her offspring's coat colour, but also that of subsequent generations. Later, Cropley et al. (2006) reported the effects of specific timing of maternal dietary methyl supplementation on the coat colour of offspring. They found that maternal supplementation only during midgestation substantially affects offspring coat colour, and this effect is inherited by the next generation, presumably through germ-line modifications during grandmaternal supplementation.

Differences in male diet in the period before mating have also been associated with altered offspring outcome. For example, males who eat a diet consisting of betel nuts prior to mating have offspring that develop diabetes and metabolic syndrome, a phenotype which can be inherited for at least three generations (Boucher et al. 1994). Males that undergo a 24-h complete fast two weeks before mating have offspring with reduced serum glucose and altered levels of corticosterone and IGF1 (Anderson et al. 2006).

Inheritance of acquired habit

There is good evidence in support of the hypothesis that the acquired habits of one generation are passed on to the next. Durken (1923) found that when the caterpillars of the common cabbage white butterfly (Pieris brassica) were exposed to orange light, they developed into pupa in a large proportion (66%) of which the white colours of the cuticle were suppressed and the green blood of the chrysalis shone through. If these pupa were allowed to develop into butterflies and these butterflies were paired, and the eggs laid allowed to develop into another generation of caterpillars, and these caterpillars again exposed to orange light, then the proportion of green caterpillars increased to nearly 100%; whereas if the caterpillars were subject to ordinary daylight still about 34% turned green, a number enormously greater than that found in nature. These experiments have been repeated by Harrison (1927) and completely confirmed. It is worth noting that Mcdougall attained world-wide fame as a brilliant experimental psychologist. He repeatedly showed that when rats were forced by unpleasant experience to acquire a habit, they produced offspring which acquired the habit more readily than did their parents until, at the end of nine years; they acquired the habit at the first contact with the unpleasant experience (Rhine & Mcdougall 1933; Mcdougall 1938).

Lindqvist et al. (2007) showed that the offspring of chickens raised in stressful conditions have an affected phenotype and brain gene expression, which mirrors that of their parents, supporting previous observations that have shown that offspring phenotypes may be affected by parental experiences preceding pregnancy, and even persist over more than one generation. Most recently, Natt et al. (2009) investigated the inheritance of acquired behaviour adaptations and brain gene expression in chickens. Parents were raised in an unpredictable (UL) or in predictable diurnal light rhythm (PL, 12:12 h light:dark). In a foraging test, UL birds pecked more at freely available compared to birds from the PL group. The female offspring of UL birds, raised in predictable light conditions without parental contact, showed a similar foraging behaviour, differing from the offspring of PL birds. Furthermore, adult offspring of UL birds performed more food pecks in a dominance test, showed a higher preference for high-energy food, survived better and were heavier than were the offspring of PL parents. These adaptations would be transmitted to the offspring. In addition, Arai et al. (2009) demonstrated that exposure of 15-day-old mice to two weeks of an enriched environment enhanced long-term potentiation not only in these enriched mice but also in their future offspring through early adolescence, even if the offspring never experienced an enriched environment.

Behavioural plasticity is the ability of an animal to modify its behaviour so as to respond to an environmental change, and imprinting is an important form of adaptive behavioural plasticity. The nematode Caenorhabditis elegans undergoes such adaptation to new environments by imprinting: attractive odorants, when present during the first larval stage, produce life-long olfactory imprints that enhance attraction and egg-laying rates in the adults (Remy & Horbert 2005). Recently, Remy (2010) reported that the olfactory imprint can be transmitted to the next generation. If the imprint is generated successively over more than four generations, it is stably inherited through many following generations, instead of just transmitting through one further generation.

Inheritance of acquired immunity

There has been a controversy about the inheritance of acquired immunity since the beginning of the twentieth century (Ogilvie 1901). Guttmann and Aust (1963) found that a C3H male mouse made tolerant of strain A spleen cells fathered a significant number of offspring which were also tolerant of A strain major histocompatibility complex (MHC) antigens. Kanazawa and Imai (1974) demonstrated that the foreign MHC antigens introduced into chimeric fathers could be found in their offspring. Later, Gorczynski and Steele (1980, 1981) found that male mice made tolerant of MHC antigens of a single strain or of two mouse strains appeared to transmit the specific tolerance to both first- and second-generation offspring. Their experiments consisted first in the induction of tolerance in male mice of the CBA strain by the classical procedure of neonatal injection of lymphoid and bone marrow cells from a hybrid donor strain bearing foreign MHC antigens. When they reached adulthood, these tolerant males were mated with normal females. Interestingly, both generations, neither of which had been intentionally exposed to the tolerogen, showed a wide range of response, from normal down to undetectable. A high proportion (50–60% of the first generation and 20–40% of the second generation) gave responses significantly below those given by control CBA mice. This finding was regarded as Lamarckian revival in immunology (Taylor 1980).

Inheritance of characters acquired by blood transfusion

As early as 1871, Galton carried out intervarietal blood transfusions experiments among rabbits, but failed to induce heritable changes. Later, Sopikov (1954) reestablished the use of this method for acquired inheritance, and showed that repeated transfusion of blood of Australorp roosters (black feathers) to White Leghorn hens, and subsequent mating of these hens with roosters of the same breed (White Leghorn) had yielded progeny of a modified inheritance. During the 1950s and 1970s, Sopikov's observations were confirmed not only by many Soviet researchers (Golubev 1966; Gromov 1970), but also by investigators in France, Switzerland and other countries. For example, Stroun et al. (1963) reported that birds of a White Leghorn strain repeatedly injected with blood from the grey guinea fowl produced progeny with some grey or black-flecked feathers in the second and later generations. Leroy et al. (1966) injected whole or fractionated blood of the guinea fowl into a strain of Rhode Island Red chickens and obtained progeny in the first and second generation with extensive changes in the quantity and distribution of melanin pigment visible in the plumage. The transmissibility of modifications was continued to the seventh generation after a single series of injections of guinea fowl blood.

Inheritance of characters acquired by a dam from her former mates (telegony)

Telegony is the effect of a previous mating on subsequent progeny. Its principle is that females are impregnated by the first males to which they are mated, so that some of their subsequent offspring, regardless of their actual father, will show influence of the first male (Rabaud 1914). In chapter 11 of his book The Variation, Darwin (1868) collected many alleged examples of ‘the direct action of the male element on the female form’. He considered it to be of special importance for understanding the mechanisms of heredity and development. Spencer (1893) regarded it as a major weapon in his debate with Weismann over the reality of the inheritance of acquired characters. However, in Morgan's (1926) ‘Experimental Zoology’, telegony was dismissed as ‘another breeder's myth’, and was used as an illustration of the ‘credulity of men who have not been trained as to the value of evidence’.

There were many alleged cases of telegony in animals (Lowe, 1896; Rabaud 1914; Zhelnin 1964). For example, when normal females that had borne at least three litters to males made tolerant neonatally were subsequently mated to normal, nonimmune males, the offspring showed a hyporesponsive phenotype that did not differ from that of the progeny of fathers made tolerant neonatally. The response of this offspring was significantly lower than the response of mice born to normal females mated only to normal mates (Gorczynski et al. 1983). This type of observation has also been reported by Sobey and Conolly (1986), who demonstrated that male domestic rabbits, mating after recovering from myxomatosis, may transmit immunity to progeny born to the female in the next seven months, including progeny sired by other males lacking immunity.

Different mechanisms underlying the inheritance of acquired characters

Explanations proposed by Lamarck and Darwin

According to Lamarck, animals respond to new needs required by the novel circumstances of a changing environment by adaptive efforts. These efforts lead to changing habits and functions and, over time, they translate into new forms and structures. The resulting enforced use and disuse are inherited. Lamarck assumed that the vectors of change within the organisms were the so-called ‘fluids’. He hypothesised that the ‘subtle fluids’, emitted by the sense organs and circulating through the nerves influenced and caused modifications of the ‘ponderable fluids’ and, ultimately, modifications of the body structures (Por 2006). Today this may seem to be a nebulous and crude hypothesis.

In his provisional hypothesis of Pangenesis, Darwin also assumed that subtle ‘gemmules’ (the embryonic form of our modern genes), passing freely through tissues and between cells, were the vectors of change within organisms. Darwin proposed that all cells of the body throw off gemmules at various developmental stages and these gemmules circulate throughout the organism and enter the sex cells. If the cells of one part of the body underwent change as the result of environmental change, these cells in that body part consequently throw off modified gemmules, which are transmitted to the offspring. That is to say, a gemmule from an organ that had been modified by the environment or by the effects of use and disuse would perpetuate that modification in the offspring. Darwin's explanation of inherited mutilations, which as he notes, occur ‘especially or perhaps exclusively’ when the injury has been followed by disease, is that all the representative gemmules, which would develop or repair or reproduce the injured part are attracted to the diseased surface during the reparative process and are there destroyed by the morbid action. The fact that mutilations were not transmitted by inheritance was accounted for by the presence of sufficient gemmules from previous generations. In this manner, Darwin accounted for the inheritance of acquired characters, of which he had no reason to doubt. Now we can affirm that Darwin's idea that gemmules are the molecular carriers of hereditary characters, they multiply by self-replication and they circulate throughout the organisms has been removed from the position of a provisional hypothesis to that of a well-founded theory. Over the past several decades, the presence of free-circulating nucleic acids in plasma and serum of healthy and diseased human beings has been a well-established phenomenon. One can gain some insights by comparing Darwin's gemmules with circulating nucleic acids (Liu 2008).

Mechanism based on epigenetic inheritance

Epigenetic inheritance is a component of epigenetics. It occurs when phenotypic variations that do not stem from variations in DNA base sequences are transmitted to subsequent generations of cells or organisms. Epigenetic effects are often heritable, in the sense that they are passed on from one cell generation to the next. Epigenetic inheritance has been known for constituting a form of Lamarckian transmission of acquired characters in animals and plants (Jablonka & Raz 2009). There are many examples (such as diet-induced heritable changes and the inheritance of acquired habit), often explained under the blanket term ‘epigenetic inheritance’, of altered genetic or environmental conditions resulting in heritable changes in gene expression. For example, in their investigation regarding the inheritance of acquired behaviour adaptations and brain gene expression in chickens, using cDNA microarrays, Natt et al. (2009) found that the differential brain gene expression caused by the challenge was mirrored in the offspring, suggesting that unpredictable food access caused seeming adaptive responses in feeding behaviour, which may have been transmitted to the offspring by means of epigenetic mechanisms, including regulation of immune genes. In the example of the viable yellow agouti mice, the dietary supplementation led to clear shifts in the phenotypes related to a concomitant increase in DNA methylation at the A ^vy locus, representing a clear demonstration of how nutrition can influence the epigenetic organisation of genes, and as a consequence, can have long-term effects on gene expression and phenotype (Feil 2006).

It should be noted that prions are an unusual form of epigenetics. Their stable inheritance and complex phenotypes come about through protein folding. It has been proposed that prions could be a mechanism for the inheritance of acquired characters (Liu 2007; Koonin & Wolf 2009; Halfmann & Lindquist 2010).

Mechanism based on RNA-mediated inheritance

Temin's stunning finding of reverse transcriptase revolutionised our understanding of genome function, because it showed the existence of the flow of information from RNA to DNA. Inspired by Temin's discovery, Steele (1979) proposed a plausible genetic mechanism (‘somatic selection’ hypothesis) whereby acquired immunity may be inherited. This hypothesis states that mutant somatic information, if selected sufficiently for expression with the soma, will be transmitted to the germline in the form of RNA captured by endogenous retroviral vectors. Once in the germline, the RNA will be transcribed to DNA by reverse transcriptase and become integrated into the germline DNA. Steele's hypothesis is regarded as a modern molecular view of Darwin's Pangenesis, and shares analogies with a recently discovered form of RNA-mediated inheritance, compatible with a Lamarckian-type adaptation (Sciamanna et al. 2009). It has been fully confirmed that RNA can act as a template for DNA synthesis in the reverse transcription of retroviruses and retrotransposons, and guide genome rearrangement (Storici et al. 2007; Nowacki et al. 2008). Many details of this mechanism have been covered in an updated review on Lamarck and immunity by Steele (2009).

Mechanism based on horizontal gene transfer

Landman (1991) suggested that genes of organisms could be divided into two groups. Most are inherited ‘vertically’ from ancestors, but some were acquired ‘horizontally’ at different times from viruses, plasmids, bacteria or other agents. He argued that the acquisition of foreign nucleic acid is not only a major mechanism for acquisition of new traits, but also an important source of organised, integrated genetic information for organisms ranging from bacteria to mammals. In recent years, horizontal gene transfer has been recognised as an important mechanism of inheritance of acquired characters and a major force in evolution (Goldenfeld & Woese 2007). Prokaryotes readily obtain DNA from the environment, with phages and plasmids serving as vehicles, and in many cases directly through the transformation pathway. The absorbed DNA often integrates into prokaryotic chromosomes and can be fixed in a population if the transferred genetic material confers even a slight selective advantage onto the recipient (Koonin & Wolf 2009).

Horizontal gene transfer may well explain the inheritance of characters acquired by blood transfusion and by a dam from her former mates. The facts of heritable changes induced by blood transfusion led Stroun and Anker (2005) to suggest the hypothesis that nucleic acids are released by living cells and circulate throughout the whole organism. Circulating nucleic acids occur ubiquitously and are bioactive in living organisms. With the discovery of circulating DNA in blood and the ability of foreign DNA being integrated into the host genome and expressed in the progeny, mechanism exists for horizontal gene transfer from one animal to another by blood transfusion. In addition, there is also increasing evidence that free foetal DNA persists in the maternal circulation after delivery of the foetus. By using quantitative real-time PCR, Lo et al. (1997) found a surprisingly high mean concentration (6.2% of total plasma DNA) of foetal DNA in maternal plasma. The discovery of circulating foetal DNA in maternal blood sheds a new light on the influence of the hybrid embryo on its mother. It has been supposed that the circulating DNA released by the hybrid embryo is transferred into the mother's body by circulation, and later is incorporated into the mother's subsequent hybrid embryo by another male, thus being able to influence the characteristics of the subsequent offspring (Liu 2011).

Conclusion

I would like to close this discourse with a quotation of Morgan (1926)'s comments on Darwin's Pangenesis: ‘The theory was proposed primarily to explain how acquired characters are transmitted. If specific changes in the body of the parent are transmitted to the offspring, some such theory is required. If the changes in the body are not transmitted, there is no need of such a theory’. Over the past several decades, a considerable body of experimental evidence has accumulated for the inheritance of acquired characters, which can be explained by different mechanisms. It still pays to go back to Darwin's Pangenesis, not only because there is good evidence for the inheritance of acquired characters but also because in many respects it is surprisingly modern. I greatly appreciate Landman (1991)'s view that observations concerning the inheritance of acquired characters are fully compatible with current concepts of molecular genetics. It has always appeared to me that the inheritance of acquired character is a necessary part of the theory of evolution, and that Darwin's Pangenesis throws more light on the inheritance of acquired characters than any other genetical theory yet advanced. It is now time to recognise that a new understanding of the inheritance of acquired characters opens a broader perspective on genetics and evolution.

Acknowledgements

We are deeply indebted to the editor and the reviewers for their valuable comments and important suggestions.