How to Construct an Organism

What I cannot create I do not understand.

— Richard P. Feynman

Not so long ago, newspaper headlines around the world proclaimed that scientists had created "artificial life." This astonishing news referred to an experiment from the laboratory of maverick molecular biologist Craig Venter, in which the DNA molecule of a simple type of bacteria had been artificially synthesized from its chemical building blocks (with some curious embellishments, like Venter's email address encrypted in the DNA's genetic code), and then inserted into a different species of bacteria, replacing that cell's own genome. Amazingly, this procedure resulted in a living bacterial cell that went on to divide and produce a colony of bacteria.

Beyond its sheer technical wizardry, Venter's experiment seems to offer a unique insight into the nature of heredity — the transmission of biological information across generations that causes offspring to resemble their parents, and can thereby enable evolution by natural selection. After all, Venter's research group had managed to decouple two fundamental components of a cellular organism — the genome (that is, the DNA sequence) and its cytoplasmic surroundings (that is, the immensely complex biomolecular machinery that constitutes a living cell). The resulting bacterial chimera, which combines the genome of one species with the cytoplasm of another, should therefore tell us something about the roles of the DNA sequence and the cytoplasm in the transmission of organismal traits across generations. Did Venter's bacterium resemble the species from which it got its DNA sequence, the species from which it got its cytoplasm, or both?

Reports on Venter's experiment emphasized the role of the genome in converting the bacterial host cell into a different species of bacteria: the genome induced changes in the features of the cell into which it had been inserted, such that, after several cycles of cell division, the descendants of the original chimeric cell came to resemble the genome-donor species. This result illustrates the DNA's well-known role in heredity: the base-pair sequence of the DNA molecule encodes information that is expressed in the features of the organism. Indeed, from here, it seems a small step to conclude that the cytoplasm (and, by extension, any multicelled body) is fully determined by the genome, and that the DNA sequence is all we need to know to understand heredity. Venter's experiment thus seems to provide a powerful confirmation of a concept of heredity that has underpinned genetics and evolutionary biology for nearly a century.

But take a closer look at Venter's experiment and the picture becomes less clear. Although many media reports gave the impression that Venter's "artificial" organism was created from a genome in a petri dish, the bacterial chimera actually consisted of a completely natural bacterial cell in which only one of many molecular components had been replaced with an artificial substitute. This is an important reality check: although it's now possible to synthesize a DNA strand, the possibility of creating a fully synthetic cell remains the stuff of science fiction. In fact, rather than demonstrating the creation of artificial life, Venter's experiment neatly illustrates a universal property of cellular life-forms: all living cells come from preexisting cells, forming an unbroken cytoplasmic lineage stretching back to the origin of cellular life. This continuity of the cytoplasm is as universal and fundamental a feature of cellular life-forms as the continuity of the genome. Of course, cytoplasmic continuity does not in itself demonstrate that the cytoplasm plays an independent role in heredity. After all, the features of the cytoplasm could be fully encoded in the genes. Yet, the potential for a nongenetic dimension of heredity clearly exists.

The continuity of the cell lineage has been recognized since the mid-nineteenth century but, since the dawn of classical genetics in the early twentieth century, many biologists have been at pains to deny or downplay the role of nongenetic factors in heredity, arguing that the transmission of organismal features across generations results more or less entirely from the transmission of genes in the cell nucleus. Genes were assumed to be impervious to environmental influence, so that an individual could only transmit traits that it had itself inherited from its parents. These ideas gained prominence while the term "gene" still referred to an entirely theoretical entity, and long before molecular biologists uncovered DNA's structure and the genetic code. More recently, this view was popularized by Richard Dawkins in his memorable image of the body as a lumbering robot built by genes to promote their own replication. But this purely genetic concept of heredity was never firmly backed by evidence or logic. Venter's chimeric bacteria were foreshadowed by late nineteenth-century embryological experiments that combined the cytoplasm of one species with a nucleus from another species, providing the first hints that the cytoplasm is not a homogeneous jelly but a complex machine whose components and three-dimensional structure control early development. Further tantalizing hints of a nongenetic dimension to heredity were provided by the work of mid-twentieth-century biologists who discovered that mechanical manipulation of the structure of single-celled organisms like Paramecium could result in variations that were passed down unchanged over many generations. Today, after many more clues have come to light, biologists are finally beginning to reconsider the possibility that there is more to heredity than genes.

RETURN OF THE NEANDERTHALS?

Venter's experiment raises intriguing questions about the nature of heredity at the level of a single cell, but what about multicelled organisms like plants and animals? A single cell's cytoplasm is divided in half each time the cell divides and then supplemented with newly synthesized proteins encoded by the genome. It is this process of gradual conversion that allowed the bacterial genome to gradually reset features of the host cell in Venter's experiment. Can such conversion also reset the features of more complex life-forms?

Consider an example at the opposite extreme of the complexity gradient — the recent idea of resurrecting a Neanderthal. Some people believe that such a feat could be accomplished by implanting a synthetic Neanderthal genome (whose sequence was recently deciphered from DNA fragments extracted from ancient bones) into a modern human egg or stem cell deprived of its own genome. Ethical considerations aside, it would be extremely interesting to compare the physical and mental traits of our enigmatic...