|Home | About | Journals | Submit | Contact Us | Français|
I seem to have been born into it. Not in that I was born into a scientific environment. My father was in the Canadian military, and I grew up going to parochial schools and moving every few years. But I was born into it in the sense that I was very good at some things and terrible at others, especially spelling. I went through school an average ‘B-student’, exceptional only in that I accomplished this without any Bs. The sciences were easy and attractive, but there was no illuminated path to a career in science. School taught us about ‘gentlemen’ and monks whose individual efforts established the foundations of modern science, but I assumed I needed to find a real career to support a hobby of doing experiments. Now, when I anguish over the difficulty of funding our work, I remember this youthful view and marvel that I get paid for what I do.
Miraculously, when we moved to Montreal, McGill admitted me, the peculiar B-student. I gained access to a wealth of intense science courses well above my head. I took far too many courses, learned an extraordinary amount, and again achieved the distinction of being a B-student without Bs. Ed Southin taught a magnificent course entirely based on the literature of the still infant field of molecular biology. My interests in physics went by the wayside – I was entirely hooked by the elegance of this science. I got into the laboratory of Barad Mukherjee, the first of many mentors who were to let me play at science unfettered by any disciplined plan. I was 18 and infatuated with science. I loved every part of it – the ideas, the insights in the literature, the experiments, the inventive part of getting methods to work, the excitement of doing new things and the thrill of seeing the outcome. After 42 years, science is still new and vibrant, and I am still in love.
Though the circumstances were complicated and the goal outrageously ambitious, the reasons were simple. As a student in Boulder Colorado, I lost my first advisor, Joe Daniel, when he left to take a chairmanship. Jacques Pène, a young faculty member working on bacteriophage gene expression, talked to me of his interest in starting to work with Volvox, a colonial green alga, as a model for developmental biology. It is a beautiful organism with a remarkably simple embryonic development and two-day life cycle. I was fascinated by the possibilities and I joined his lab to begin this project. I quickly discovered that it was easy to isolate mutations with altered morphology. I wanted to dissect the connections between gene and morphology. At the time (1971), there was one success along these lines. The newly introduced Laemmli SDS gels resolved most of the proteins produced by bacteriophage, allowing investigators to dissect the progression of gene expression during phage infection and assess the affects of mutations. Because I could synchronize Volvox development and could isolate mutants, I thought that I could use the same experimental paradigm; however, I needed a gel system that could resolve the much larger number of proteins made by a eukaryote. I set my goal at squaring the resolution of SDS gels by combining two methods in a two-dimensional separation – this eventually gave me the ability to detect about 2,000 gene products.
Hardly! There were a million things that needed to be worked out and a million things that could go wrong – and most did. I had a curious blend of stubbornness, confidence and optimism that led me to believe that I knew exactly what was needed to improve the separation no matter the failures and the ugly looking results along the way. It was a consuming process.
Perhaps you could say that it gave me the first of many lessons in scientific humility. Establishing a new genetic model has many components and all that I had was lots of mutants with cute phenotypes. It seemed that the testing grounds for the new method ought to be in a system with people and tools that might help me. I chose Escherichia coli, where I could address questions that had been well delineated by others. For example, I was able to show that all detectable effects of cyclic AMP are mediated by the one known receptor. The questions and approaches were those of ‘proteomics’, although the work presaged the name and the era when the scientific community developed a taste for large-scale analysis.
At the time it seemed natural and appropriate. I lost my second advisor when Jacques Pène left for a position in New Jersey. So, in the midst of developing the method, I moved to the laboratory of David Hirsh, a young Assistant Professor studying the nematode Caenorhabditis elegans. When I wrote the paper, it didn’t really occur to me that I should include any other author. While the sole authorship still seems to be a fair reflection of my autonomy, I feel bad that it does not acknowledge the debt that I owe my advisors for their support, for the confidence they placed in me, and for their willingness to just let me go. Additionally, there was a terrific interactive environment in the Department in Boulder, and I owe a great deal to Larry Gold and members of his laboratory, among others, for technical help and for enthusiastic promotion of the work when the method proved so powerful.
I wanted to understand the molecular code that biology used to shape an embryo and I wasn’t getting there by analysing changes in protein expression. Stimulated by the environment at UCSF and smart friends, I started looking at fly genetics for ways to approach development. Ed Lewis’s papers convinced me that identifying homeotic genes would provide the needed clues, and I thought the new recombinant DNA biology would provide the requisite tools. We set out to clone homeotic genes thinking of it as a lonely adventure, only to discover that we were participants in an exploding field.
Science changes remarkably over the period of a scientific career, but scientists tend not to change – many retire working on a topic inherited from an advisor during their training. Science is supposed to be about innovation, but increasingly, fear for career security seems to channel scientists into ruts. So, yes! Not only do I advise people to change, I train them to look for new questions and encourage them to find the boldness to pursue the adventure that science should be.
Yes and the difficulties are of two sorts. First, it is difficult in much the way a vigorous physical work-out is difficult. It takes an abundance of scholarly effort to develop a foundation of knowledge in a new area. It also takes deep thought to identify novel directions – after all it would be pointless to simply switch to things that others are already pursuing. Like exercise, these efforts come with rewards. Learning new things is refreshing and a new area turns one’s thinking from little intricacies to the big picture.
The second type of difficulty, unfortunately, is burdensome, and sadly it is imposed by one’s peers. We don’t allow each other to change. Instead of instilling confidence in our abilities, our scientific reputations are the chains that bind us to the research question that we are identified with.
Personally, identifying new problems has been rewarding, and rejuvenating, but there are costs and limitations. Not every effort to start something new takes off, and stumbling can take more energy than identifying a clear path. Additionally, doing something important and making an impact takes time, so starting in a new area is not an annual affair. I have usually invested at least ten years into the pursuit of a particular area. On balance, I think changes in research focus have added importantly to what we’ve been able to accomplish, and has enriched the research careers of the people I’ve trained. Really, my only regret is that with each change of research area, there has been a change in scientific colleagues, and I find that I am not able to change my affiliations and affections as readily as my science.
Patrick O’Farrell is a Professor of Biochemistry and Biophysics at the University of California, San Francisco (UCSF). As a graduate student in Boulder Colorado he developed two-dimensional gel electrophoresis of proteins. He moved to UCSF for postdoctoral work with Gordon Tomkins and subsequently with Bruce Alberts. In 1979 he started his own laboratory at UCSF, and became one of the pioneers of molecular analysis of embryonic pattern formation. He identified the gene encoding the homeodomain protein Engrailed and his analysis of its function in segmentation defined processes used throughout the regulatory cascade that patterns the embryos. He then turned to a study of the control of cell division during development. In recent work, he has begun to explore mechanisms that integrate mitochondrial function with the rest of the organism.