| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print October 12, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
San Raffaele University, Chromatin Dynamics Unit, Milan, Italy
1Correspondence: San Raffaele University, Chromatin Dynamics Unit, via Olgettina 58, 20132 Milan, Italy. E-mail: bianchi.marco{at}hsr.it
This work was selected as a Pivotal Advance because the paper shows that HMGB1, also co-discovered as amphoterin, in a fully purified state does not activate Toll-like receptors 2 and 4, but nevertheless still has considerable DNA-binding capacity, chemotactic effects, and developmental activities.
Dr. Rauvala, you were the first to purify an extracellular protein that, as it turned out, is also a nuclear protein. How did you come across amphoterin?
HR: In the mid 80s we were searching for neurite outgrowth promoting factors that would act on cortical neurons. This had been my major research goal for several years, since I had established my own lab after a postdoctoral experience in the US in 1979–1980. Not many such factors were known, and the most interesting were fibroblast growth factors (FGFs). They were shown to regulate neuronal development, and to bind strongly to heparin and heparan sulfates. We homogenized young rat brains, and fractionated extracts on a heparin column, where not many proteins would bind at high salt concentration. The assay was simple: we adsorbed protein fractions on tissue culture plates, seeded primary rat brain neurons on the plates, and watched for neurite extensions after 24 h.
In fact, the project went very well: we obtained a good amount of a protein that migrated as a single band of 30 kDa in Coomassie gels, which was obviously different from FGFs and was very active. We called this protein p30, and we realized very early on that it binds tightly to cation exchange columns, and thus must contain many basic residues. This worried me a bit, since cationic proteins were known to be very common in the nucleus, where they bind DNA. So we raised antibodies against p30 and used them and lactoperoxidase-catalyzed cell surface iodination to label intact neuroblastoma cells and brain neurons. This satisfied me that p30 was expressed in the extracellular space, essentially as a membrane-associated protein.
The purification of p30 was reported in The Journal of Biological Chemistry in 1987 [1 ]. The name amphoterin came later, when we realized that p30 had both positively and negatively charged segments.
What was your personal reaction, and that of your colleagues, when you realized that amphoterin and HMGB1 were the same protein?
HR: As I said, we had generated antibodies and had produced evidence that p30 was a surface protein. In 1988, we published these data in The Journal of Cell Biology, together with the sequence of the first 11 amino acids of p30, which were determined by Edman degradation [2 ]. Using an antibody against the first 11 amino acids, we had also cloned a cDNA from a lambda library of rat brain, and had gotten the partial sequence of the protein. Neither the protein nor the cDNA sequences were present in the rudimentary databases at the time, and we were sure that p30 was a novel protein. There was no signal peptide for secretion in the cDNA, but this did not worry me: it was already known that FGFs lack a signal peptide but can be secreted.
However, on the very day JCB accepted the paper for publication, someone in my lab came excited to my office and showed to me that in fact the sequence of p30 matched a sequence in the EMBL database, that of High Mobility Group 1 protein (HMG1, as it was called at the time). There was no paper yet, but the author for the sequence was you, the interviewer. I remember I had a very difficult day. In the following weeks, I read all the vast published literature on HMG1, and it clearly showed that it was a DNA-binding, nuclear protein. In the end, we added a note to the proofs of the JCB paper, saying that the sequence of p30 matched that of HMG1, but that the adhesive properties and the surface localization of this protein suggested that extracellular mechanisms should be considered as the basis for the role of p30 in neurite extension.
The real problems came after the paper was published in December 1988. Most people believed that p30 was really an artifact, and the editors of the journals to whom we submitted our subsequent papers gave us a cold shoulder. An internal review within the University was negative, and suggested that our work on p30 should be terminated. It was a very, very difficult time for me, and my lab could not have survived if we did not have another project on a different adhesive and neurite outgrowth promoting protein, heparin binding growth-associated molecule (HB-GAM).
Finally, we found a very critical but balanced and supportive editor of The Journal Biological Chemistry. He decided that, despite the criticisms of the reviewers, our data showing that HMG1 and amphoterin, which were identical chemically but had different functions, were sound. We had generated several different affinity-purified antibodies, covering the whole of the protein, to show that sequences belonging to HMG1 were also present on the surface of neurons. It took us three years to publish those data [3 ].
For some time, I was personally unaware of your research. We were totally absorbed by HMGB1’s nuclear functions [4 ]. You named the protein differently, and I did not know what you knew—that the cDNAs were identical. When your JBC paper with the cDNA sequence came out, I thought that although HMGB1 had neurite outgrowth promoting properties, these were probably biologically irrelevant—the protein was clearly nuclear and nuclear only, in the cells we used. I was too busy trying to survive as a young independent associate professor in Italy to even think about trying to reproduce your immunocytochemistry on neural cells. Even after inviting you to give a seminar in Milan, I was not fully convinced. I am happy I was never involved in reviewing any of your papers or grants; otherwise I might have been rather dismissive. When did you eventually feel vindicated?
HR: The turning point came when Ann Marie Schmidt’s group identified a receptor for amphoterin/HMGB1 in 1995 [5 ]. They were looking for a ligand for the receptor of advanced glycation end products (RAGE), and they purified peptides whose sequence corresponded to amphoterin. I did not know this group and their research, but this was great news that I read in JBC. Only extracellular proteins can have receptors. I was totally reassured, even though the skepticism of others abated only gradually.
To me, the great eye-opener came with Kevin Tracey’s 1999 discovery that HMGB1 can be secreted by macrophages. I remember I read his Science paper [6 ] on a July afternoon, and I decided it was worth trying to determine whether the various pure HMGB1 preparations we had in our freezer had any inflammatory properties. Moreover, we knew that HMGB1, contrary to most chromatin proteins, was not tightly associated to DNA and could diffuse away upon cell permeabilization. And Paola Scaffidi, a new graduate student in my lab, had just shown that apoptotic cells retained HMGB1 bound to the nuclear remnants. After that, it was a downhill race for us as well as for everybody else. And now you are showing that fully purified HMGB1/amphoterin does not promote the secretion of inflammatory cytokines, but only has chemoattractant and cytoskeleton remodeling properties. This is fully consistent with the role you propose for it in neural development. So, what will you be doing next with amphoterin/HMGB1?
HR: We are still interested in the same problem: the development and plasticity of neural connections. We are trying to do structural studies on the RAGE-HMGB1 interaction. We have also identified syndecan as another receptor for HMGB1, and are studying syndecan’s function. Moreover, we are pursuing some other proteins, including the AMIGOs (amphoterin-induced gene and ORF). We are also doing some work on zebrafish: you have knocked out HMGB1 in mice, but morpholinos in zebrafish can reveal additional information.
The Finnish government is investing very decisively in research, in contrast to several other European governments. You are the director of the Neuroscience Center. How do you judge your country's efforts, and what are the general prospects for you and Finnish science in general?
HR: Yes, Finland is investing a lot in science and technology. Of course, most of the investment is in electronics and telecoms, because of the Nokia effect. In fact, Nokia is the only global success that Finland has had so far, and the government is trying to extend this success, and to replicate it in other fields, including biology and biotechnology. This is very good for us, and there is a very positive atmosphere. There has also been talk of a Bio-Nokia, since neural cells and silicon chips do similar operations: they compute. But the timescale of this work is too long, and so far Nokia has not been investing in neuroscience. To be fair, biotechnology in Finland is only in its infant stage, and there are only initial (but encouraging) successes.
Can you tell us a bit about the person who was a major influence in your professional life, and how this person helped you to choose the course your career has taken?
HR: Certainly. I was most influenced by Johan Järnefelt, a professor of Medical Chemistry whom I met during my MD studies. I was very interested in psychology, and wanted to be a psychiatrist. Professor Järnefelt taught us that every process in our bodies, even in the brain, had to be rooted in biology and chemistry, and that the biochemical processes underlying physiology and pathology were becoming accessible. Molecular biology had not yet been applied to medical studies in the late 60s, but nonetheless, he showed that biochemistry could be very informative. I joined his research group and worked on gangliosides. After I got my medical degree, I saw patients to earn my keep, and often worked in the lab for free during the rest of the time. There were very few salaries then for doing medical research. We have come a long way since, but at the same time, medical students are showing less interest in basic research. This is a mistake, and I am doing my best to demonstrate to them how intellectually rewarding research can be.
|
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
