Henry over at Webiocosm reminded me that parasitism appears over and over again in the living world. The cycle of defenses and counterdefenses drives an evolutionary spiral between host and parasite, sometimes even resulting in mutual dependence and benefit.
Recently this evolutionary perspective has been brought to bear on a superficially different phenomenon-- the masses of repetitive DNA, such as Alu repeats , which infest the human genome. Alu repeats, together with other repeat families, make up a staggering 40% of the genomic sequence, without coding for any genes. A recent article in Nature Genetics shows that Alu repeats and their compadres form interactions within the human genome resembling a jungle-like ecosystem. One family of repetitive elements, the LINEs, consists of a complete parasite package, with special DNA sequences flanking a minigene which encodes the enzyme necessary for making more copies of itself. The much smaller Alu repeats take the freeloading one step further, and appropriate the LINE recognition sequences, and therefore are parisitic on the parasites. The Alu repeats are present in the human genome to the tune of a million or so copies. Quite an ecosystem!
One feature to keep in mind with the parasite analogy is that, numerous as they are, these repeats do not increase in a single healthy human. Instead, a new Alu repeat is inserted on average in 1/200 live births. There are statistical indications that Alu repeats are also removed over time, possibly in the male germline, but on the whole, it is a very slow interaction relative to, say, fleas on a dog.
There are two costs to humans with carrying all these repeats. Stable repetitive DNA sequences must be copied along with the rest of the genome with each cell division, which costs a lot of energy. The host might prefer to put this energy other uses (think about a tapeworm with a third of your weight). Secondly, during the less frequent instances when a new copy is made, it may get inserted into an important genetic region, resulting in disease. Both these effects have been seen in fruit flies. More of these repetitive elements correlates with fewer hatched eggs . Thus, as with all parasites, their "success" ends up hitting a wall when the well-being of the host starts to suffer.
However, specific instances have been found of Alu actions on genomic function (I wrote about two examples here and here . In both cases, the location of an Alu repeat ends up influencing gene expression or gene repair). So this set of freeloaders has indeed influenced human evolution.