The discovery of the molecular mechanism of the neuroparalytic action of tetanus and botulinum neurotoxins by Cesare Montecucco
My scientific interest in presynaptic neurotoxins begun during a semi-sabbatical at the Institut Pasteur in 1984 in the lab of Jean-Pierre Changeux and was stimulated by discussion with Patrice Boquet, also at Pasteur. On my return to Padova, this interest was shared by Giampietro (Gipi) Schiavo who joined the lab about that time: so young, so intelligent, so talented, with a solar character that illuminated the oldish building of the Institute of General Pathology of the University of Padova. He was not scared by our poverty and by the fact that we were doing everything the hard way: building most of our instruments with our hands, including gel apparatuses, bacterial culture shakers, borrowing salts and reagents from our wealthier neighbors.
Our science was full of readings, almost no congresses and many talks,. The best ones were those mademany of them on Sundays spent walking and climbing with any weather in the Dolomites. Our, and of experiments reacheding all directions as we were always ready to learn novel techniques and to collaborate with colleagues. Right from the beginning the central questions of the mechanism of action and of the evolutionary origin of tetanus and botulinum neurotoxin were a preferred subject occupied most of our talks. We first tackled the former question by producing large amounts of super pure tetanus toxin to be used in crystallization experiments in the beliefd that crystals would have unveiled, at the same time, the atomic structure and the activity of the toxin. Anaerobic cultures were run in secret over week ends with open windows to disperse the terrible smell that would have condemned us. We tried every possible condition and method of crystallization for years with no success. And it is no consolation that nobody has, as yet, crystallized tetanus toxin.
Then in 1986, the groupp lead by Heiner Nieman working at the University of Giessen in Germany and Neal Fairweather with Val Lyness at Wellcome Research Laboratories (UK) reported the entire sequence of tetanus toxin. We inspected it carefully and concentrated on an unusual histidine-rich stretch in the middle of what was known to be the as the aactive chain. However we were unable to decode the hidden message contained by employing sequence comparison bioinformatics programs that were becoming available: the result always was carbonic anhydrase! The secret concealedhidden in the histidine rich string was unveiled upon reading a paper of Neil Rawlings and Alan Barrett (Biochem. J. april 1991) describing a novel class of metalloendopeptidases characterized by the signature HExxH. I remember running up from the ground floor library to the lab, where we had been meanwhile joined by Ornella Rossetto as a graduate student. Journals could not be moved out from the library, but Gipi and Ornella were rapidly convinced that tetanus and botulinum neurotoxins could be metalloproteases and we begun experiments to prove this hypothesisby the presence of the mere presence the HExxH fingerprint in the middle of the L chain !
We were trained as chemists and biologists and the experiments that came up to our mind were firmly based on chemistry and biochemistry. We simply followed the experimental schedules with no weekend interruptions. Meanwhile, we began to read papers on synaptic vesicles, neurotransmitter release and electrophysiology of the neuromuscular junction. In fact we were looking forward We wanted to identify the target(s) of these metalloproteases and we wanted to win the competition. Within few months we produced solid evidence that tetanus and botulinum neurotoxins A, B and E (BoNT/A, /B and /E) were metalloproteases. BoNTs were provided to us by Bibhuti Das Gupta of the University of Wisconsin-Madison with whom we had previously established a long-lasting collaboration which became then a friendship.
Having obtained this first major result, within a short, but very intense and passionate, period of time we found targets and sites of cleavage of all seven BoNT serotypes and of tetanus toxin, in a strong competition with the groups of Jahn and Niemann. Their target were VAMP/synaptobrevin, SNAP-25 and syntaxin. In 1993 we showed that an animal species can be resistant to tetanus by mutation at the toxin cleavage site and in 1994 we provided the first evidence that tetanus and botulinum neurotoxins recognize and cleave their three substrates (VAMP/synaptobrevin, SNAP-25 and syntaxin) via an interaction with an exosite located on the protein substrates at a distance from the cleavage site. We termed this a double interaction mode of recognition. Later on work of Brunger’s and Swaminatham’s labs showed that BoNT-substrate interaction actually involves multiple exosites.
At this point it was clear that tetanus and botulinum neurotoxins caused paralysis by removing a large part of the three SNARE proteins thus preventing the assembly of the SNARE complex, proposed by J. Rothman, and therefore the release of neurotransmitter. More in general, our findings provided the strongest evidence that the three toxin substrates were essential components of the nanomachine which mediates the Ca2+-activated release of neurotransmitters and other molecules including hormones. It still remains to be explained the exact molecular mode of action of BoNT/A, the toxin used in human therapy, because it only removes few residues from the carboxyl terminus of SNAP-25.
This was the end of a fabulous pioneering period, but, sadlyworse, it also was the end of the mountain climbing and of the conversations with Gipi, as he moved to Jim Rothman’s lab in New York. For months the lab seemed empty to me, and though later on the lab obtained other relevant results on Helicobacter pylori virulence factors, on anthrax toxins, on snake neurotoxins and other subjects, including again the clostridial neurotoxins, I never experienced any more the intellectual and emotional intensity of that magic triennium 1991-1994, for I am so grateful to life.