NOT having any family is tough. Often unappreciated and uncomfortably different, orphans have to fight to fit in and battle against the odds to realise their potential. Those who succeed, from Aristotle to Steve Jobs, sometimes change the world.
Who would have thought that our DNA plays host to a similar cast of foundlings? When biologists began sequencing genomes, they discovered that up to a third of genes in each species seemed to have no parents or family of any kind. Nevertheless, some of these 鈥渙rphan genes鈥� are high achievers, and a few even seem have played a part in the evolution of the human brain.
But where do they come from? With no obvious ancestry, it was as if these genes had appeared from nowhere, but that couldn鈥檛 be true. Everyone assumed that as we learned more, we would discover what had happened to their families. But we haven鈥檛 鈥� quite the opposite, in fact.
Advertisement
Ever since we discovered genes, biologists have been pondering their origins. At the dawn of life, the very first genes must have been thrown up by chance. But life almost certainly began in an RNA world, so back then, genes weren鈥檛 just blueprints for making enzymes that guide chemical reactions 鈥� they themselves were the enzymes. If random processes threw up a piece of RNA that could help make more copies of itself, natural selection would have kicked in straight away.
As living cells evolved, though, things became much more complex. A gene became a piece of DNA coding for a protein. For a protein to be made, an RNA copy of the DNA has to be created. This cannot happen without 鈥淒NA switches鈥�, which are actually just extra bits of DNA alongside the protein-coding bits saying 鈥渃opy this DNA into RNA鈥�. Next, the RNA has to get to the protein-making factories. In complex cells, this requires the presence of yet more extra sequences, which act as labels saying 鈥渆xport me鈥� and 鈥渟tart making the protein from here鈥�.
The upshot is that the chances of random mutations turning a bit of junk DNA into a new gene seem infinitesimally small. As the French biologist Fran莽ois Jacob famously wrote 35 years ago, 鈥渢he probability that a functional protein would appear de novo by random association of amino acids is practically zero鈥�.
Instead, back in the 1970s it was suggested that the accidental copying of genes can result in a single gene giving rise to a whole family of genes, rather like the way animals branch into families of related species over time. It鈥檚 common for entire genes to be inadvertently duplicated. Spare copies are usually lost, but sometimes the duplicates come to share the function of the original gene between them, or one can diverge and take on a new function.
Take the light-sensing pigments known as opsins. The various opsins in our eyes are not just related to each other, they are also related to the opsins found in all other animals, from jellyfish to insects. The thousands of different opsin genes found across the animal kingdom all evolved by duplication, starting with a single gene in a common ancestor living around 700 million years ago (see diagram).
Most genes belong to similar families, and their ancestry can be traced back many millions of years. But when the yeast genome was sequenced around 15 years ago, it was discovered that around a third of yeast genes appeared to have no family. The term orphans (sometimes spelt ORFans) was used to describe individual genes, or small groups of very similar genes, with no known relatives.
鈥淚f you see a gene and you can鈥檛 find a relative you get suspicious,鈥� says , who studies the evolution of complex traits at Penn State University. Some suggested orphans were the genetic equivalent of living fossils like the coelacanth, the last surviving members of an ancient family. Others thought they were nothing special, just normal genes whose family hadn鈥檛 been found yet. After all, the sequencing of entire genomes had only just begun.
So many orphans
But as the genomes of more and more organisms were sequenced, genetic family reunions proved to be the exception rather than the rule. Orphan genes have since been found in every genome sequenced to date, from mosquito to man, roundworm to rat, and their numbers are still growing.
The study of orphan genes is still in its infancy, and we know very little about most of them. Those we do know about are a mixed bag. Some are involved with the repair and organisation of DNA, or in controlling the activity of other genes. The insect orphan flightin, which encodes a muscular wing protein, evolved to aid flight. And in a study published last year, Manyuan Long of the University of Chicago and his team showed that two recently evolved young insect orphans help shape foraging behaviour in the fruit fly Drosophila.
In corals, jellyfish and polyps, orphan genes guide the development of explosive stinging cells, sophisticated structures that launch toxin-filled capsules to stun prey. In the freshwater polyp Hydra, orphans guide the development of feeding tentacles around the organism鈥檚 mouth. And the polar cod鈥檚 orphan antifreeze gene enables it to survive life in the icy Arctic.
However improbable鈥�
Curiously, orphan genes are often expressed in the testes 鈥� and in the brain. Lately, some have even dared speculate that orphan genes have contributed to the evolution of the biggest innovation of all, the human brain. In 2011, Long and his colleagues identified 198 orphan genes in humans, chimpanzees and orang-utans that are expressed in the prefrontal cortex, the region of the brain associated with advanced cognitive abilities. Of these, 54 were specific to humans. In evolutionary terms, the genes are young, less than 25 million years old, and their arrival seems to coincide with the expansion of this brain area in primates. 鈥淚t suggests that these new genes are correlated with the evolution of the brain,鈥� says Long.
Critics argue that most genes, new or old, are somehow involved with the workings of the brain, and that correlation does not prove causation. But Long cites a recent animal study that lends credence to the theory. Expressing one of the human orphan genes, SRGAP2C, in the neurons of developing mice doesn鈥檛 make the animals鈥� brains grow bigger. But it does encourage the nerve cells to grow denser arrays of dendritic spines, the tiny protrusions that enable neurons to connect with their neighbours. Having more connections might increase computing power, he argues. So these recently evolved human genes may have helped shape the human brain. 鈥淚 think we鈥檝e underestimated orphan genes,鈥� says geneticist Diethard Tautz of the Max-Planck Institute for Evolutionary Biology in Pl枚n, Germany.
But where did they come from? In 2003, Tautz and a colleague suggested that orphan genes are formed by duplication, but then evolve so rapidly that any similarity to the original is obliterated. And they did have evidence that seemed to support this idea. They showed that than non-orphans.
Orphan genes were thus crowbarred into the old model of genes arising by duplication. Later studies, however, suggest this can only explain the origins of a minority of orphans. So while the process is clearly important, it鈥檚 not the whole story. 鈥淭he idea seemed reasonable at the time,鈥� says Tautz, 鈥渂ecause the alternative seemed so unlikely.鈥�
鈥淭he idea seemed reasonable at the time because the alternative was so unlikely鈥�
The alternative? The only other possibility was that genes really can evolve from scratch, from random stretches of non-coding DNA. This is the idea long dismissed as completely implausible, because the leap from non-coding DNA to a gene with a useful protein product was considered so huge as to be impossible. But nature hasn鈥檛 read the textbooks. A few years ago, evidence began to emerge of genes created 鈥渄e novo鈥� in yeast, rice, mice and fruit flies. Then in 2009, David Knowles and Aoife McLysaght of the University of Dublin in Ireland showed that three orphan genes in people had indeed been created from scratch.
They found that DNA sequences nearly identical to the genes existed in several other primates but were non-coding, meaning the genes must have arisen sometime after the human-chimpanzee split. They also showed that the orphan genes are transcribed to RNA and then translated into protein in multiple tissues, though the functions of the genes are not known yet.
In 2011, another team . McLysaght thinks this might be an overestimate 鈥� she believes that de novo gene synthesis is a rare phenomenon.
Some other researchers, however, . A study of 270 primate orphan genes, led by M. Mar Alb脿 and Macarena Toll-Riera of the Municipal Foundation Institute for Medical Research in Barcelona, Spain, found that only a quarter could be explained by rapid evolution after duplication (). Instead, around 60 per cent appeared to be new. 鈥淒e novo evolution is clearly a strong force 鈥� constantly generating new genes over time,鈥� says Tautz. 鈥淚t seems possible that most orphan genes have evolved through de novo evolution.鈥�
But how can it be possible? Knowles and McLysaght showed that the orphan genes they found sit next to and slightly overlap existing, older genes, so the orphans might be able to 鈥渂orrow鈥� their switches. Similarly, Alb脿 and Toll-Riera found that half of the 270 primate orphans had acquired sequences from the genes of 鈥渢ransposable elements鈥�, genetic parasites that can jump around in the genome. Meanwhile, the ENCODE study of the human genome published earlier this year showed that our DNA is littered with millions of potentially useful short switching sequences, and that single switches can interact with many genes.
All this suggests that it is relatively easy for non-coding DNA to acquire the switches needed for RNA copies of it to be made. Indeed, the ENCODE study found that as much as 80 per cent of DNA is copied into RNA at least occasionally. Some argue that all this RNA is functional, but another interpretation is that most of this activity is just noise, and that junk DNA is routinely transcribed into RNA.
Proto-genes
If so, we are basically experimenting with thousands of potential new genes all the time 鈥� and Anne-Ruxandra Carvunis of the University of California, San Diego, has shown that this is indeed the case, at least in yeast. Last year, her team analysed 108,000 short, unknown but potentially protein-coding sequences in the yeast genome (). More than 1000 were interacting with the cell鈥檚 protein-making factory, suggesting that they were being converted to proteins. 鈥淭hese may just be the tip of the iceberg,鈥� says Carvunis.
Her findings suggest that the protein-making factories in yeast are constantly churning out new proteins, allowing them to be 鈥渢ested鈥�, and she suspects that same is happening in all complex organisms. In between non-coding DNA and fully fledged genes, Carvunis thinks there is a whole continuum of 鈥減roto-genes鈥�. Most will code for proteins that are neutral or harmful, so there will be no selection and the vast majority of proto-genes will revert to non-coding DNA sooner or later. But a few proto-genes that are neutral or maybe even helpful will sometimes persist, and start to gather beneficial mutations. Over millions of years of natural selection, they can become a proper gene 鈥� and thus is an orphan born.
鈥淏etween non-coding DNA and fully fledged genes there could be a continuum of 鈥榩roto-genes鈥欌��
All this helps explain why orphan genes are often expressed in the testis. In most cells, DNA is tightly packaged, which reduces the chances of RNA copies being made. In certain immature sperm cells, however, the structure is more open, making it easier for proto-genes to be copied into RNA. Over time, the gene may come to be expressed in other tissues and evolve new functions.
New discoveries about the nature of proteins also make the idea of genes arising de novo seem far more plausible. It was once thought proteins must be folded into a delicate, precise 3D structure to work properly, but it now seems many exist in a state of intrinsic disorder, flitting through thousands of different possible conformations, all the while remaining perfectly functional. About half of human proteins have at least one long intrinsically disordered segment, while 10 per cent are disordered from beginning to end.
Peter Tompa of the Flanders Institute of Biotechnology in Brussels, who studies intrinsically disordered proteins, suspects that new orphan genes are likely to code for disordered proteins because they are easier to make than folded proteins. And disordered proteins often play a role in cell signalling and regulation. 鈥淚 wouldn鈥檛 be surprised if orphan genes turn out to have regulatory functions,鈥� says Tompa.
Perhaps this helps explain why orphan genes can become essential very quickly. In 2010, Long鈥檚 team used RNA interference to switch off evolutionarily old and new genes in fruit flies. They found that new genes, including orphans, were just as likely to be essential for life as old genes (). 鈥淭his goes against the textbooks, which say the genes encoding essential functions were created in ancient times,鈥� says Long.
We still have a lot to learn about orphan genes, but we are now starting to trace their ancestry. And it looks as if we couldn鈥檛 find the families of most orphans because they don鈥檛 really have families. The raw DNA from which they sprung can be traced, but as genes they are the first of their kind. In this sense, the term orphans may be a misnomer. Perhaps they ought to be renamed Pinnochio genes 鈥� non-genes carved by chance and natural selection into proper, 鈥渓iving鈥� genes.
