NAME: John Postlethwait
POSITION: Professor, biology, University of Oregon Institute of Neuroscience
Visiting scientist, Imperial Cancer Research Fund, Oxford University — 1989-1990
Visiting scientist, CNRS, Laboratory of Eukaryotic Molecular Genetics — 1982-1983
Associate professor, biology, University of Oregon — 1977-1981
Assistant professor, biology, University of Oregon — 1971-1977
Postdoc, molecular genetics, Harvard University — 1970-1971
PhD, developmental genetics, Case Western Reserve University — 1970
BA, biology, Purdue University — 1966
Earlier this month, a team of researchers from the University of Oregon published data in Nature Genetics indicating that a particular microRNA, miR-140, plays an important role in zebrafish palatogenesis and that its over-expression leads to cleft palate.
While these findings might offer new insights into the causes of human craniofacial deformities, they also suggest that miRNAs may have an evolutionary significance as a fine-tuning mechanism for skull morphology, according to the authors.
This week, RNAi News spoke with John Postlethwait, senior author on the Nature Genetics paper, about the data.
Let’s start with some background on your lab and your research focus.
Our focus really is in developmental genomics; we’re interested in how genomes change during evolution, particularly with respect to the organization of embryonic development. The development of the skeletal system is one of the things we study … and [that’s] what led to [the work published in Nature Genetics].
We had discovered a number of years ago that there was a whole-genome duplication in the zebrafish lineage, so we’ve been interested in how the duplicated genes from that whole-genome duplication — which is shared by all teleost fish — have evolved and how they might be important in making different [features in related] fish.
How did the idea to look at microRNAs come into this?
A paper came out from [Erno] Weinholds’ group [at the Hubrecht Laboratory] in the Netherlands [detailing how] they went through 120 or so microRNAs in zebrafish and made beautiful in situs to look at where and when each of these microRNAs is expressed during development.
I looked through those in situs in their database and picked out a number of [miRNAs] that were expressed specifically in the developing bones and cartilage of the face. One of these was microRNA-140. I asked Xinjun He, who is one of the two lead authors on the [Nature Genetics] paper, to take a look at this microRNA, see if he could verify the expression, and then do some functional analysis.
He injected the microRNA into zebrafish embryos and found that the embryos that resulted after this injection had a cleft palate. That would suggest that the microRNA was blocking the expression of some gene that is essential to prevent cleft palate. But when you look for potential targets bioinformatically, you get just an enormous number of potential targets, so we were wondering where to go from there.
In a kind of miraculous moment, the following happened: Every Monday morning, we have our local zebrafish group meeting. … Johann [Eberhart, a postdoc who is the other lead author on the Nature Genetics paper], gave a talk about a zebrafish mutant he had found … that looked exactly like Xinjun’s animals.
We thought, “Whoa. Here’s this mutation. This gene that Johann had mutated might be a target [for miR-140].” After the meeting, Xinjun looked in his list and, indeed, it was a target with a couple of binding sites for miR-140.
That then initiated investigations to look at the 3’ UTR of this mutation that Johann had found, which is in the [platelet-derived growth factor receptor alpha] gene. Then Xinjun took the 3’ UTR, hooked it up to a GFP reporter, and showed that the miR-140 does inhibit the translation of GFP, specifically when that 3’ UTR is attached to the GFP. If the miR binding site is gone from that 3’ UTR, there is no effect, showing us that this microRNA truly is controlling the expression of Pdgfra.
Since improper activity of this microRNA leads to cleft palate, do you have any idea what its role might be when it inhibits that gene normally?
I think so. The cells that form the zebrafish palate, or [human] palates, come from neural crest, which are cells at the dorsal part of the brain. The ones that form the palate migrate in front of the eye and down into the place where they make the palate. In doing so, they have to migrate past a particular set of neurons that lead from the eye in order to get to the place where the palate needs to be: in between the mouth and the brain.
The cells migrate there because there is a source of growth factor that continues to move as the animal develops. The source of the growth factor, a protein, moves down from the top of the eye, down in front of the eye, and down below the eye. The neural crest cells are making the receptor for the growth factor, the Pdgf, [which] causes the cells to move towards the source of the growth factor.
The miR-140 slows that migration and regulates how many cells get to the place of the palate. Different species have different shaped palates, and what we imagine — this hasn’t been proven yet — is that the relative levels of miR-140 are important in different species for controlling how many cells get down there to form the palate.
In humans, the Pdgfra gene has the binding site for miR-140 and the molecular/genetic system is just the same [as zebrafish] — it has these cells that are migrating to the region of palate formation and so on. The whole situation is set up so that one could imagine that if there is an upset in the amount of Pdgfra that is made due to either the over-expression or under-expression of miR-140, then you can get the inappropriate palate formation in a human baby.
Have you also looked at what happens when you remove miR-140?
Yes. That also disrupts palate development. We used morpholinos … directed against miR-140 … and that also causes a problem in the migration because the cells now have too much Pgdfra, too much of the receptor, and they kind of get stuck and don’t get all the way down to the palate.
In either case, you don’t get enough cells to the palate. If you have too much, then the neural crest cells can’t [sense the source of the growth factor and migrate to the site of the palate]. If you’ve got too much Pdgfra, then they get stuck because they are binding too much.
Do you plan to see if this is happening in mammals?
Not in my lab. We don’t do mouse embryology. But there are a couple of papers on miR-140 in mice, but they’re not in vivo — they’re not looking at developing embryos, they’re looking at cell cultures, but they have shown that [the miRNA] affects skeletogenesis. They’ve also shown that the mouse miR-140 has a similar expression pattern to that in zebrafish in developing mouse embryos.
But nobody has done a functional analysis; it’s much more difficult to do the functional analysis in mouse embryos than in zebrafish embryos.
Are there experiments that your lab will conduct to build on this work?
What I’m interested in is this evolutionary situation. Everybody says that microRNAs are important for evolution, but I am not aware of any case in which someone has shown that a specific microRNA is expressed in a different time, amount, or place in two different species or two different populations of the same species and causes the difference that has evolved between those [animals].
For example, the medaka fish has a palate that is shaped very differently from the zebrafish. What we hypothesize is that miR-140 may be at least partially responsible for this difference in palate shape that has evolved since in those two species since their last common ancestor.
We’re investigating, then, medaka fish and stickleback fish to see if miR-140 may be modulating the patterning of the skeletal elements for the palate to see if it could be responsible for evolutionary change in response to natural selection between these species. To my knowledge, nobody has actually done that kind of thing for any microRNA.