It is useful, at times, to look back and contemplate how you arrived in your current position.  Some of what brings you to a particular place comes of your own design.  The rest of the direction may result from circumstance.  In the world of science, design and serendipity often mingle to produce fascinating results.  As an aspiring scientist you might be taken by a particular topic or personality you find fascinating.  As you carry on, obtain experience, dabble or even master new technology, science is changing along with you.  By the end, or even at the middle of your journey, you may find yourself in an interesting place you never expected to be.

As a young college student at UC Berkeley, Yi-Hsin Liu had an interest in immunology.  Twelve years later, as a research assistant professor at the CCMB, Dr. Liu conducts a research program that focuses on the role a gene called Msx-1 has on the development of cranial sutures.  Just how did he get here from there? 

While keeping his eye on potential Ph.D. advisors at Berkeley, Liu secured his first laboratory position through an independent study’s course working in microbiologist Dr. Denis Ohman’s laboratory on the virulence of Pseudomonas bacteria, which at that time was thought to play a role in causing cystic fibrosis.  It wasn’t exactly what one would classify as an immunology lab.  Rather, it served as the time when Liu learned how to do experiments, hands-on, at the bench.  Liu analyzed a Pseudomonas gene promoter that was responsive to mutating ultraviolet light.  I remember I did thousands of CAT reporter assays.  It was the first time I learned how, sometimes, redundant science can and has to be, recalled Liu.

Liu was still thinking about immunology as he finished up at Berkeley and prepared to apply to graduate school.  One academic institution took a particular interest in having Yi enroll as a graduate student.  That university was USC.  Liu ultimately decided to attend USC and joined the Microbiology department.  He did a series of rotations, first through a lab that studied hepatitis B gene regulation, then through a lab that looked at the cancer related oncogenes, Jun and Fos and lastly through a lab that analyzed the patterning Hox genes.  He learned how to do gene regulation experiments and tissue culture and, what would later prove to be an influential technique in his repertoire, transgenic animal technology. 

Liu settled in Dr. Chi Minh Nguyen-Huu’s lab to complete his PhD project.  Nguyen handed down a difficult project to Liu that everyone in the lab had been reluctant to tackle, even the senior graduate students. Nowadays, scientists typically mutate a specific gene to see what role it plays in development or disease progression.  The project Liu had been given had been derived with the opposite approach- almost a reverse genetics.  Nguyen’s group had randomly inserted a reporter gene into a mouse embryo’s DNA and ended up with a mutation that caused lethality.  The problem was that they didn’t know what gene they had interrupted.  It was a mouse mutant that my mentor wanted to characterize to see how the transgene insertion was causing the lethality in the mouse.  Being a new graduate student I was very naïve.  I knew nothing about histology or pathology. So I set up to clone that gene instead.  It turned out to be a very difficult project, explained Liu. 

Liu used different approaches to fish out the unknown gene that was creating the phenotype in their mouse.  He analyzed samples using the transgene as a probe to look for differences in tissue expression, with no luck.   He dug out a large section of DNA near the site of the transgene insertion, 25,000 base pairs, and then chopped it into small pieces so he could use those pieces as probes, still no luck.  He looked at genes that had sequence homology to sections of their 25,000 base pair segment and used those as probes for the gene. Still, nothing came out. 

Then something completely unexpected and tragic occurred, Liu’s mentor passed away.  It was a blow to the entire lab and it resulted in the parceling of the group into different departments.  Liu went to the developmental biology laboratory of Rob Maxson in the department of biochemistry to finish his Ph.D.  Clearly my project didn't go on as planned by my previous mentor.  But Rob was able to seize on the techniques and technology I had learned in Chi's lab, explained Liu.

Maxson had recently cloned a gene called Msx-2.  Virtually nothing was known about the gene, not its function, or its regulation. At the time, making transgenic mice targeting a specific gene was anything but trivial.  So at the beginning, Liu tackled the regulation of the Msx-2 gene.  By resolving some of those questions Liu and Maxson hoped to glean information indirectly about the function of the Msx-2 gene.  We started examining the promoter activity in cultured cells because it's normally the easiest approach.  But it turned out that none of the cell lines would work, even the ones that naturally had high levels of Msx-2 expression, remembered Liu. 

Neither Liu nor Maxson realized it at the time, but the negative data was a signature of Msx-2 function and not of finicky cells.  Later, as Liu explains, they were able to demonstrate that the Msx2 gene shuts itself down under certain circumstances.  Basically, if you insert an extra copy of Msx-2, it shuts itself down.  It has a negative feedback loop, explained Liu.  But that understanding came only in hindsight. 

Toiling for months with little progress, Liu and Maxson decided to explore the experimental route they had initially passed over, the generation of transgenic animals.  At the beginning it was difficult because the technology wasn't up to speed.  There were so many variables, animals, injections, needles, surgical procedures, hormones, reimplantation, he explained.  Eventually Liu’s efforts with the help of lab coleagues, Frank Sangiorgi and Nancy Wu, met with success. His first transgenic Msx-2 mouse was made so that the expression of Msx-2 could be visualized during the course of development.  Low and behold the expression patterns were very interesting.  Msx-2 was expressed in many different places and the most interesting areas were cranial neural crest related. Cranial neural crest cells act as progenitor cells that provide numerous cell types including the face, bones, teeth, and neurons.  Liu’s work was pulling him into the world of craniofacial biology.

Liu finished his Ph.D. working on the characterization of Msx-2 and continued as a post-doctoral research associate with Maxson.  His commitment to staying in Maxson's lab was due, in part, to the discovery that a particular cranial neural crest cell lineage, one that gave rise to bones of the skull, was affected by the Msx-2 mutation.  Maxson, Liu and CCMB senior researcher, Malcolm Snead, had established a collaboration with geneticist, Mimi Jabs at Johns Hopkins University in a search for gene targets responsible for human craniosynostosis, a congenital condition where the skull bones prematurely fuse.  Msx-2 had become a candidate gene for the condition because of its localization to craniofacial tissues.  The group discovered a linkage between the Msx-2 gene and Boston-type craniosynostosis.  The next question, as Liu explained, was what change to Msx-2 was causing the condition?  After sequencing DNA from affected patients, the researchers determined that the disease was the result of a single base pair mutation to Msx-2 in a region that was responsible for DNA binding.

The discovery of the human mutation made the project very attractive to me.  It's basically why I stayed for the post-doc, said Liu.  Maxson and Liu went on to generate a transgenic mouse that closely resembled the human condition.  While reproducing the condition was an important demonstration, it served an even greater purpose to Liu and Maxson.  It gave them the opportunity to study the aspect of bone formation affected by the Msx-2 mutation.   "When Msx2 is overexpressed it accelerates bone formation.  It does so by increasing the proliferation rate of the pre-osteoblast in the suture region," explained Liu, "So the mutation increases the number of active osteoblasts in the suture which leads to increased bone deposition." 

In 1996, Liu was approached by Snead to come to the CCMB as a research assistant professor.  The Center was looking for a 'close insider', someone to manage a center grant project that had been left behind by then director, Hal Slavkin, who was leaving to head the NIH's NIDCR.  It served as an opportunity for Liu to work more independently, to mentor students and technicians, and to write grant proposals to fund his own studies.   "Initially I wanted to continue to work on the Msx2 promoter, but I needed to develop my own niche."   Liu and Maxson continue to collaborate on Msx-2 projects together, but Liu's independent projects came to focus on Msx-1, a gene with close similarity to Msx-2.  

Initially, scientists believed that Msx-1 was a homologue of Msx-2, that the two proteins could be used interchangeably during development.  But that assumption only holds true under certain circumstances.  As Liu describes, "They're highly redundant in the developing limb.  That is, Msx-1 and Msx-2 determine whether the interdigital zone will undergo apoptosis or not.  If you delete both of these genes, the cells won't die and you get webbed fingers and toes."

In the craniofacial region it is an entirely different scenario.  The two proteins are functional antagonists to one another, explains Liu, "Msx-2 potentiates bone formation while Msx-1 inhibits it."    Liu's fledgling group is trying to determine the gene pathways Msx-1 regulates to exert such an effect over bone formation.  Their results are still preliminary but what they believe is that Msx-2 and Msx-1 regulate bone formation via separate pathways.  "We have obtained a Msx-1 knock-out mouse and have generated specific transgenic animals to selectively express Msx-1 in different cell types in the skull.  Their hunch is based on the observation that, as Liu explains, "Msx-2 basically alters expression of an important growth factor receptor, the FGF receptor, in the cranial region while Msx-1 does not."  In this region of the cranium, he continues, Msx-1 and Msx-2 are decidedly not equivalent.  Msx-1 does not affect cell cycling in the bone cells and thus does not play a role in their proliferation.  Instead, Msx-1 may play a role in maintaining suture cells in undifferentiated state.

What's different about conducting research as a student versus a post-doc versus a young faculty member?  As a faculty member, there's a need to create and maintain a "viable research program" says Liu.  Research is not just about having a reasonable scientific concept to work on, there are additional stakes in the endeavor.  "You are not responsible for only yourself anymore," explains Liu. One has to manage grant money and employees. Faculty need to think about job security not only for themselves but for their students and technicians as well. These issues aside, says Liu, "My main priority is to enjoy my work and contribute useful information to the scientific community and society."

A selection of recent publications: 

• Liu YH, Snead ML, Maxson RE Jr (2000). Transgenic mouse models of craniofacial disorders. Methods Mol Biol. 137: 499-512. 

• Sun X, Lewandoski M, Meyers EN, Liu YH, Maxson RE Jr, Martin GR (2000). Conditional inactivation of Fgf4 reveals complexity of signaling during limb bud development. Nat Genet. 25(1): 83-6.

• Dhamija S, Liu Y, Yamada Y, Snead ML, Krebsbach PH (1999). Cloning and characterization of the murine ameloblastin promoter. J Biol Chem. 274(29): 20738-43.

• Dodig M, Tadic T, Kronenberg MS, Dacic S, Liu YH, Maxson R, Rowe DW, Lichtler AC (1999). Ectopic Msx2 overexpression inhibits and Msx2 antisense stimulates calvarial osteoblast differentiation. Dev Biol. 209(2): 298-307.

• Liu YH, Tang Z, Kundu RK, Wu L, Luo W, Zhu D, Sangiorgi F, Snead ML, Maxson RE (1999). Msx2 gene dosage influences the number of proliferative osteogenic cells in growth centers of the developing murine skull: a possible mechanism for MSX2-mediated craniosynostosis in humans. Dev Biol. 205(2): 260-74.

• Liu YH, Ma L, Kundu R, Ignelzi M, Sangiorgi F, Wu L, Luo W, Snead ML, Maxson R (1996). Function of the Msx2 gene in the morphogenesis of the skull. Ann N Y Acad Sci. 785: 48-58.

• Snead ML, Paine ML, Chen LS, Luo BY, Zhou DH, Lei YP, Liu YH, Maxson RE Jr (1996).  The murine amelogenin promoter: developmentally regulated expression in transgenic animals. Connect Tissue Res. 35(1-4): 41-7.

• Liu YH, Kundu R, Wu L, Luo W, Ignelzi MA Jr, Snead ML, Maxson RE Jr (1995). Premature suture closure and ectopic cranial bone in mice expressing Msx2 transgenes in the developing skull. Proc Natl Acad Sci U S A. 92(13): 6137-41.