A question of heredity

BIOLOGISTS firmly believe that parents cannot transmit to their children the traits they acquire during their lifetimes. Though a language can be learnt and passed on to the younger generation, when it comes to heritable traits, the Darwinian theory of evolution remains the cornerstone of existing knowledge of evolution of life. So, if muscles are developed by exercise, that does not imply that the progeny is likely to grow up to be more muscular than other children. A recent paper by Irmgard Roemer and her team from Berlin and Cambridge raises the startling possibility that certain acquired traits may indeed be inherited (Current Biology, Vol 7, No 4).

The starting point of Roemer's study was the observation that in two strains of laboratory mice (named B and D for short), certain nuclear transplantation experiments between the strains were causing alterations in patterns of gene expression. Three factors play a role in this. First, the genome of the developing embryo, which is surrounded by an internal environment, the cytoplasm, derived from its mother. Second, the genetic material derived from the mother via the nucleus of the egg. Andfinally, an equal complement of genetic material coming from the nucleus of the father's sperm.

When specimens from either of the strains B and D are bred within the strain, the resulting embryo can be denoted by the symbols BBB or DDD, in accordance with the three factors of maternal cytoplasmic environment, maternal genes, and paternal genes. A hybrid formed by mating a B mother with a D father would be BBD, while one derived from a D mother and B father would be DDB, as two of the three influencing factors come from the mother. In all these cases, as also in other conventional crosses between a hybrid and either parental type, the pattern of proteins synthesised by the embryo is what one might term normal.

In striking contrast, two proteins are strongly repressed in both BDB and DBD hybrids. These are artificially generated embryos and can never occur in nature. For example, BDB implies that a D-type female nucleus is surrounded, not by its own cytoplasm as is normally the case, but by B-type cytoplasm. In order to generate such a 'nucleocytoplasmic, hybrid, the nucleus from a D egg has to be removed and transplanted into a B egg, after the removal of its own nucleus. Curiously, the two proteins with depressed levels of expression in BDB and DBD hybrids have interrelated functions. One belongs to the class of major urinary proteins (mup) and the other is an olfactory marker protein (Omp). omp is expressed in the mucosal layer in the nose and transported to the olfactory bulb at the front of the brain. There, in combination with molecules such as mup, it is involved in 'presenting' pheromones, or sexual attractants, to the brain.

What this goes to show is that early environmental influences - in this case the environment provided by the cytoplasm of the egg - can have long-lasting implications on gene expression. Also, given the nature of the affected proteins, the effect appears to extend to adult behaviour (though this has not been tested).

What was more important was that the altered patterns of gene expression found in nucleocytoplasmic (NC) hybrids continued into the next generation. This was true even when an NC hybrid male was crossed with a B or D-type female. The significance of this lies in that the cytoplasm of the newly fertilised embryo, which contains some genetic information within structures called mitochondria, is derived almost entirely from the mother. However, the only genes that an Nc hybrid male can pass on are those that are contained in its nucleus. Therefore if an Nc hybrid male can transmit changed patterns of gene expression, it signifies that a characteristic acquired early in life was successfully passed down in a hereditary fashion.

The researchers have suggested an explanation that accords with modern Darwinian theory. It was observed that in all cases in which an animal showed a reduction in the level of expression of the mup protein, there was a concomitant change in the mup gene: it had an increased number of methyl groups attached to the DNA.

Studies of many systems over a number of years have shown that, firstly, methylation Of DNA can affect the level of expression of the protein encoded in that molecule. Secondly, the level of methylation, which can be influenced by external cues, can also persist over generations. In other words, a changed pattern of gene expression can occur, not only because of a mutational change in the conventional sense, but because of a reversible chemical modification (in this case, methylation) of the DNA. And if the modification does not revert in the immediately succeeding generation, the change can persist. Indeed, ' there is an example in which methylation increases stepwise from one gener, ation to the next, and leads to a steady generational increase in the severity of the effects it causes.

The results indicate that the environment within which early embryonic development takes place could have long-term, transgenerational effects. This should make scientists wary of assuming that in vivo fertilisation or embryo cloning will have no long-term effects. The findings support the belief that even seemingly mundane causes have the potential to harm future generations.