Time travel may be possible, at least in a laboratory -- researchers have reconstructed a now-extinct vertebrate gene that split into parts 500 million years ago. During the study
, published this week in Developmental Cell
, researchers replaced two mouse genes with a synthesized "ancestral" version that contains parts of both.
"The most striking finding was that we succeeded in 'simplifying' the mouse genome," said Petr Tvrdik
, lead author of the study. "We could replace two genes that are obviously essential by a single, bifunctional gene."
Tvrdik and his coauthor Mario Capecchi
, both at the University of Utah in Salt Lake City, focused on the mouse Hoxa1
genes, the two most significant descendants of the ancient vertebrate Hox1 gene
. The Hox family of genes
, all of which have between two and four paralogs in modern vertebrates, play an essential role in embryo development.
paralogs help to shape the central hindbrain -- mice lacking Hoxb1
are deficient in motor neuron function and can't perform certain facial movements such as blinking. Mice lacking Hoxa1
have a defective breathing mechanism, and die soon after birth. In humans, Hoxa1
mutations sometimes occur; victims survive into adulthood, but suffer from impaired eye movements, deafness, delayed motor development, and breathing difficulties.
Researchers had previously hypothesized
paralogs encode similar proteins and differ mainly in their regulatory sequence. To test this in the case of Hoxa1
, Tvrdik and Capecchi swapped the protein-coding regions of the two genes. The mice turned out to be normal. "We did find subtle differences, but the mice were happy with swapped genes," Tvrdik said. "Intrigued by this, we asked if we could go even further in constructing a single gene that would perhaps do both of these seemingly separated functions in the mouse brain."
switch on around the same time in a mouse embryo. But due to differences in the regulatory sequence, one remains active only for about a day, while the other continues to act for several more days. Since Hoxa1
's protein can substitute for that of Hoxb1
in the brain, Tvrdik and Capecchi hypothesized that if Hoxa1
's activation pattern could be suitably modified, it could take over Hoxb1
's function as well -- essentially reverting to the ancient Hox1
gene. To achieve this synthesis, the researchers inserted Hoxb1
's autoregulatory element, a 107 bp sequence, into Hoxa1
's promoter. In mice with Hoxb1
knocked out, two copies of this composite gene restored viability and normal facial movements. "Rather strikingly, the mouse with the bifunctional Hoxa1
gene was fine even in the absence of the other gene," Tvrdik said.
Based on their results, the researchers propose a new form of gene therapy, where a small regulatory fragment from a defective gene could be used to activate a similar, but functional, gene. This might require much less DNA transfer compared to conventional gene therapy since only a small regulatory fragment needs to be inserted into the second gene, Tvrdik said.
"Perhaps the most astonishing finding from this work is that mice can survive with normal neurological functions given one functional Hox1
paralog in place of the Hoxa1/Hoxb1
pair, Jonathan Eggenschwiler
of the department of Molecular Biology at Princeton University said in an Email. However, Eggenschwiler, who was not involved in the paper, added that even though mice with a single version of Hox1
appear normal, "slight positive selective pressure in favor of two diversified Hox1
paralogs may only be evident over long evolutionary timescales."
The study highlights the importance of the cis
-regulatory elements in the divergence of gene function, said Steven Potter
of the Developmental Biology department at the Cincinnati Children's Hospital Medical Center, also not a co-author. The elegant and novel feature of the study, according to Potter, is that the authors combine the cis
-regulatory elements from different paralogs to create a multifunctional Hox gene. "That was a bold experiment," he said. "I probably would have never tried it, because I wouldn't have thought it could work."
Links within this article
P. Tvrdik and M. Capecchi, "Reversal Hox1
gene subfunctionalization in the mouse," Developmental Cell
, 11:1-12, August 2006.
C. Holding, "Hox: Total knockout," The Scientist
, July 18, 2003.
S. Blackman, "Old genes, new tricks," The Scientist
, August 28, 2003.
J.M. Greer at al., "Maintenance of functional equivalence during paralogous Hox gene evolution," Nature
, 403(6770):661-5, Feb 10, 2000.