I've been intrigued by the elegance and the complexity of how gene expression is regulated since I was in High School in Milan, Italy. When I went to university, I wanted to do something both intellectually stimulating and potentially useful. Getting into science to do biomedical research seemed to be the best way to do this.
During my graduate studies at the Universita' degli Studi di Pavia, I started to work in Professor Silvano Riva's lab on a protein involved in RNA processing. It was during my submersion in this work that I became interested in RNA splicing and the way in which it relates to human diseases.
RNA splicing is an essential part of gene expression. In order to be used by the cell to produce proteins, the genetic information that is stored in the DNA needs first to be transferred to an intermediate molecule, chemically similar to DNA, called RNA. One of the most fascinating aspects of how this information flows is that RNA needs to be substantially modified in the process.
The most dramatic part of this processing step is the removal of most of the original RNA (or, pre-mRNA, which is a letter-by-letter transcription of its DNA template) in large internal chunks, called introns. The remaining parts, or exons, are joined together to produce a mature messenger RNA (mRNA). The process of removing the excess material interspersed within a gene to obtain a product translatable into proteins is called RNA “splicing.”
It is a process to some extent analogous to film-making, with which it even shares some of the terminology. When filming (transcription), one shoots hours of material (the pre-mRNA), of which the majority is “junk” (the introns) and needs to be discarded. Only selected scenes (the exons) are accurately cut and spliced together (splicing) to produce the final cut (the mRNA), creating a meaningful end product.
The analogy continues. Even with the same initial material, scenes can be added or removed, or alternative ones can be split in to change the message of the movie (the protein function) in a subtle or dramatic way. And if the splicing is not done correctly — for example, if splicing causes a frameshift, or if parts are missing or are unintentionally included — the end product could be drastically different, and possibly meaningless.
Coming to America — Cold Spring Harbor Laboratories
Coming out of grad school, I was really fascinated with the idea of doing my postdoctoral work in the U.S., where the opportunities and the resources to do great science are of a different league. I knew I wanted to work on splicing, so once I decided to make the move, I contacted some of the major labs in the field.
I eventually decided on Cold Spring Harbor Laboratory (CSHL), in Cold Spring Harbor, Long Island, home to one of the leading authorities in the field, Dr. Adrian Krainer. I found my time there to be inspiring. It is a world-renown institution both for the level of the science produced and for the CSHL Meetings, which regularly bring a who's-who of scientists from a variety of different fields. That can be very stimulating. It's a great opportunity to hear and meet great scientists and to put a face to the names you know from scientific papers.
Although I was always interested in both the mechanistic aspects and the applied part of splicing in relation to disease, I didn't have one specific targeted research goal. An interesting problem with splicing is that only a fraction of the apparently good potential splice sites are ever used.
The question becomes what exactly defines a splice site and why some sites are used and others are not. Furthermore, some of the sites are used in some cellular contexts, but not in others, in a regulated process known as alternative splicing.
Alternative Splicing
By alternative splicing, a single pre-mRNA can be spliced in a variety of ways to generate proteins that have similar but not identical functionality. Since the human genome was deciphered in 2000, and it was discovered that there were significantly less genes than expected, it's been determined that a good portion of the gene variability comes from alternative splicing.
It's a common phenomenon, and upward of 60 percent of genes have multiple splicing isoforms. The splicing patterns can be very complicated, and some genes can have hundreds of different isoforms (the most extreme example has over 38,000).
“ The connection between splicing regulation and cancer is a largely unexplored field, which could be very fruitful. “
Luca Cartegni, Molecular Biologist
Initially, I was involved in a project aimed at understanding how splicing factors participate in the generation of specific isoforms. Gradually, I became more interested in some of the disease aspects of the field. Elements that regulate splicing are a very common target of mutations: up to 30 to 50 percent of single-point mutations associated to genetic diseases affects splicing in one way or another.
Splicing is an easy target because mutations that hit splice sites or regulatory elements don't affect just a single amino acid, but all the ones coded by the involved exon (or all the ones downstream when a frame-shift is involved), therefore exerting a much more dramatic effect on the encoded protein.
Exonic Splicing Enhancers — Spinal Muscular Atrophy
In particular I've been focusing my work on one class of elements that regulate splicing, called Exonic Splicing Enhancers (ESE). Similar to transcriptional enhancers, these motifs work as specific binding sites for splicing factors that promote splicing by recruiting the splicing machinery. We used in vitro experiments to match the different splicing factors to their functional sites on the RNA, and then we developed a computational approach to identify such enhancers and to predict whether mutations may disrupt them.
What had begun as a side project and had morphed into my main research focus received further definition when I decided to study spinal muscular atrophy (SMA), a very severe neurodegenerative disease that affects young children. While it is not very common — occurring in one out of every 6 to 10,000 births — it is the leading genetic cause of death for children. And, as of now, there is no effective therapy.
What we know about the molecular biology of the disease has been discovered only in the last few years, since the responsible gene was identified. It's called SMN1 — survival of motor neuron 1 — and produces a ubiquitous protein that has an essential function in any cell. For reasons that are still not clear, the dosage is particularly important in neurons, so that's where the phenotype appears. There is a second gene, called SMN2 which has the potential to encode a protein identical to that of SMN1.
However, because of a splicing defect, it mainly generates a defective product. This is very important because all SMA patients have active copies of SMN2, and therefore they retain the ability to generate full length SMN protein. It's a sort of built-in gene therapy system: if you could somehow “convince” the SMN2 gene to splice correctly, than you would have a possible SMA treatment.
A New Approach
We first tried to understand what was the nature of the SMN2 splicing defect. Capitalizing on our previous studies, we demonstrated that an enhancer is lost in SMN2, and therefore splicing is inefficient. The next questions were: What can be done about this loss? Can we mimic the activity of the splicing factor that is unable to bind with something small and specific enough to represent the basis for a therapy?
While some labs are using standard high-throughput screens to try to find leads, we took a more mechanistic, basic-science approach. We exploited our knowledge of the splicing mechanism to design a small chimeric compound that acts as specific synthetic splicing factor and that promotes correct SMN2 splicing in vitro. This was a simple but novel idea and we are now developing this technology, which we called ESSENCE (which stands for Exon-Specific Splicing ENhancement by small Chimeric Effectors) to see if it can be the basis of a therapeutic approach.
What is really exciting is that this kind of approach has very broad potential applications, not only because splicing defects are common disease-associated mutations, but also because it can be used to investigate splicing mechanisms and can be applied towards the control of endogenous alternative splicing events.
As I mentioned, alternative splicing is a common and key aspect of gene expression, and its correct regulation is paramount to maintain a cell's “normal” identity. In fact, aberrant splicing is very frequently observed in tumors, and in many cases specific isoforms play a determinative role in tumor progression or in a cancer cell's ability to develop resistance to standard treatments.
Next Chapter: SKI
To go back to the film-making analogy, it is like if somebody locked themselves in the editing room and started re-cutting the movie not according to the Director's instructions but to promote their own agenda. Why did they do it and, above all, can we fix it?
The connection between splicing regulation and cancer is a largely unexplored field, which could be very fruitful. What is the exact role that the misregulation of splicing plays in many cancers? Will understanding alternative splicing in tumors lead to the identification of new targets? Can we intervene pharmacologically on the splicing events themselves to induce the production of anti-tumorigenic isoforms, or to make other treatments more effective?
These are some of the questions that I'd like to address, and Sloan-Kettering — with its resources, its emphasis on the translational aspects of research, and the opportunity it provides to collaborate with some of the best scientists in cancer research and treatment — is the perfect place to find these answers.
