Fred Sanger - gone

Fred Sanger double Nobel laureate for Chemistry died on 18th November 2013, three days ago.  It staggers me that it didn't make the news with sufficient TaRa that even I, in this news-benighted backwood, heard about it. At 95 he was the same age as James Lovelock, so his death isn't surprising but it's still a shock when when a scientific superstar dies. Neither of these men ever threw shapes like a superstar, and Sanger retired to his modest English garden in 1983. I wrote a trib to Francois Jacob in May when that key figure in molecular biology and genetics passed on.  Jacob and Monod made a key contribution by working out how the on-and-off of genes are controlled.  Sanger was more infrastructural, developing methods for making sense of biological macromolecules by sequencing them.  In a broad-brush-strokes metaphorical sense, bio-sequences are codes determined by the order of their constituent building blocks.  DNA is made up of four bases A T C G and long specifically-ordered strings these bases are translated into proteins which are made up of 20 amino acids.  Sanger's contribution was to work out nifty and then niftier ways for determining the order of the constituents. His second, niftier, way of sequencing DNA was used to sequence the human genome and millions of other sequences from thousands of different organisms all through the 80s 90s, and 00s. It was so useful and so idiot-proof that even I was able to use it to sequence a couple of hundred DNA bases in about 1986.  For this Sanger shared one half of one half of the 1980 Nobel chemistry prize with Wally Gilbert who had invented another clunkier method for achieving the same aim.

Sanger's method for sequencing DNA started the data deluge which we are still trying to keep from drowning in.  But I'd rather cite him for his 1958 Nobel for determining the sequence of the hormone insulin. This was a task requiring dogged biochemistry.  Dogged but also remarkably creative as he invented new tricks for abstracting the information and also reading the literature to hear what novel techniques he might mobilise for his project.  As functional mammalian insulin is only about 50 amino acids in two separate chains this is a finite puzzle but far from trivial.  Sanger's task was to reduce a mammoth combinatorial problem - how are these 50 AAs are ordered? (there are more than 10^30 possible answers) - into a number of smaller problems. The first step was to do a total digestion of the protein chains so that he had a countable heap of amino acids A=10, C=6, D=2 etc. Then he was able to use particular enzymes like trypsin and partial acid hydrolysis to cut the larger sequence into manageable chunks.  With a combination of paper chromatography and running the fragments out along an electric gradient he was able identify characteristic finger-prints for each bit. He also invented a way to tag the last amino acid in a chain with a bright yellow chemical and then breaking the chain into smaller constituent parts so that he could then identify what sort of amino acid that last one was. He did this first with the two natural insulin chains and then with the fragments he had created by chemical and biochemical degradation.  By integrating the position of blobs on filter paper and a catalogue of the conditions under which glycine and phenylalanine and the other amino acids acquired their yellow label, he was able to match and braid the fragments of info-string into two coherent sequences.

It took him and a pair of graduate students 10 years (!) and a lot of hard thinking but he cracked the puzzle. In doing so he showed that proteins are not amorphous lumps of amino acids but that the constituent AAs are added one-a-time in a regular, predictable order.  That's something we all know now but 65 years ago it was much less obvious-to-all-thinking-people. That was in 1952.  The knowledge that proteins exist with characteristic linear sequences laid out the biochemical landscape so that Crick and Watson could follow a philosophical beaten path to determine the similarly linear structure of DNA 10 years later.