In my laboratory we study molecular mechanisms that control formation of the normal and malformed kidney. Why bother? Well, kidney malformation is the major cause of kidney failure. Moreover, it is increasingly thought that the number of nephrons (kidney filter and attached tubular segments) that form in utero is an important determinant of adult-onset high blood pressure and kidney disease. As a paediatric nephrologist, these issues really trigger my imagination. What are the molecules that guide the complex development of kidney structures? How are these molecular pathways disrupted during kidney development? What causes these disruptions? What can we do about it?

Back in the 1980s, when I chose a postdoctoral fellowship after my clinical training in paediatrics and paediatric nephrology, the science in kidney development didn’t match the huge progress in cell and molecular biology – those were the days of cDNA cloning and reverse genetics and I wanted to be part of it. Kidney development was still being investigated using classic embryologic approaches such as tissue recombination. And so, I joined the laboratory of Bjorn Olsen, the Hershey Professor of Anatomy at Harvard. Dr. Olsen was busy cloning a multitude of collagen genes and developing a totally new understanding of how extracellular matrices are constructed. I spent over four years learning how to culture different types of glomerular cells, clone collagen genes from cDNA libraries, study collagen expression, and try to purify collagen proteins.

During the final days of my fellowship, I attended a seminar given by Oliver Smithies on gene targeting in mice. As I listened, I realized that the world would never be the same. Shortly thereafter, the first knock-out mice were published including for the Wilms’ Tumor 1 gene, which featured a major kidney phenotype. Around the same time, the opportunity to join SickKids was presented. My decision to leave Boston for Toronto was accompanied by a decision to leave the field of extracellular matrix, which I found to be rather descriptive and somewhat ‘boring’ (sorry ECM enthusiasts!). I came to Toronto ready to start an independent research program in kidney development, hoping to harness what I had learned in the Olsen lab to study kidney development.

But, how to focus my new research? Just in the nick of time, I heard another seminar in which the Bmp7 KO mouse was described – it had a major kidney phenotype but the functions of BMP7 on specific aspects of kidney development were difficult to discern. Through a new association with Jeffrey Wrana, I decided to harness my knowledge of kidney cell culture and use recombinant BMPs to study tubule morphogenesis in vitro, identify BMP receptors and investigate intracellular BMP signaling in kidney tubule cells. Our data, which revealed that BMP signaling was far more complex than originally proposed, set the stage for work we are doing to this day using transgenic mice with BMP receptor deficiency in specific kidney cells at particular stages of development. 

There is an old saying, “go where the science takes you…as long as it is a smart choice!” Our BMP work taught us how to analyze kidney development in transgenic mice at the level of morphologic, cellular and molecular events. Some of our most interesting data showed us that BMPs interact with the canonical WNT pathway in the embryonic kidney and that control of these interactions is critical to normal kidney development. Indeed, we then realized that canonical WNT signaling is highly perturbed in malformed human kidney and, thus, we have studied WNT signaling in both normal and malformed mouse embryonic kidney. We never expected to study canonical WNTs and beta-catenin but the data took us there and it is important to human kidney malformation.

Another wise piece of advice in science is to “let your trainees go with their curiosity…at least until you think they should stop!” One of my past postdoctoral fellows, Ming Hu, schmoozed with members of the Hui lab and was interested in Hedgehog signaling. Ming realized that kidney development was very abnormal in mice with Sonic Hedgehog signaling and decided to figure out why. Six years later, we are investigating numerous facets of Hedgehog signaling in the kidney and the cerebellum, all because Ming was curious and generated interesting data that got me interested. In so doing, we have discovered that HH signaling controls early kidney patterning, nephron formation and the function of specialized pacemaker cells that control coordinated contraction of the ureter. To our delight, many of our findings relate to human kidney malformed phenotypes and raise the hope that by pursuing these issues we can further understand human disease.

Twenty-five to 30 years ago we didn’t have the tools to approach kidney development at the molecular level. Since then, lots of tools have been generated and we, as beneficiaries, have identified mechanisms that control nephron formation, tubule branching, and coordinated contraction of the renal pelvis and ureter. As we learn, we identify more approachable questions, the answers to which are likely to provide more understanding of normal and malformed kidney development. Some of what we have learned can now be used as a basis to translate knowledge to patients. Like the developmental biology that gave rise to our data, translation will require development of new tools and strategies. And that is a path we are currently following, using mouse genetics, human genetics, and kidney organoids in complementary ways to understand the significance of human genetic variants, interrogate kidney biology and pathobiology, and generate human cells and tissues for regenerative purposes.