Michael Smith’s invention of a technique for introducing mutations at specific sites in genes made it possible to identify DNA sequences responsible for regulating expression of genes and amino acid sequences responsible for functional properties of proteins. Site-directed muta-genesis also raised the possibility of curing hereditable diseases of humans by correcting the defective DNA sequence in somatic or germ cells. For his achievements as DNA’s “mutagenius,” Smith shared the 1993 Nobel Prize in Chemistry.
Early Years Rowland Smith was a market gardener in Marton Moss, near Blackpool in Lancashire. He married Mary Armstead (known as “Molly”), who worked as a bookkeeper, in 1926. Michael Smith was their first child. He was educated at the local Church of England school, St. Nicholas, then at the private Arnold School for Boys, where he excelled in mathematics and science. On graduating from the Arnold School in 1950, Smith won a scholarship from the local education authority that enabled him to study chemistry at the University of Manchester. He obtained an upper-second-class degree in 1953 and began studies toward a PhD under the supervision of Dr. H. Bernard Henbest. Smith’s project involved studies on the synthesis of steroids and cyclohexane diols.
On completing his PhD in 1956 Smith took up a National Research Council postdoctoral fellowship and moved to the British Columbia Research Council Laboratories in Vancouver, Canada, to work with Har Gobind Khorana. This opportunity only became available when the award was declined by the student to whom it had originally been offered. Khorana, Smith’s postdoctoral supervisor, had trained with the nucleic acid chemist Alexander Todd at Cambridge. In the late 1940s Todd’s laboratory had achieved the complete chemical synthesis of nucleotides, the building blocks of nucleic acids. In Vancouver, Khorana was trying to develop the means of synthesizing oligonucleotides—short artificial nucleic acids—and Smith was put to work on this project.
Synthesizing DNA Phoebus Levene in the 1930s conclusively demonstrated that DNA is a deoxyribonucleotide polymer in which the bases—adenine, cytosine, guanine, and thymine—are attached to the 1’ position of deoxyri-bose and the sugars are linked to one another by phosphodiester bonds. As the furanose (5-membered ring) form of deoxyribose only has hydroxyl groups at the 1’, 3’, and 5’ positions, it was clear that the phosphodiester bonds must involve the 3’ hydroxyl of one nucleotide and the 5’ hydroxyl of the adjacent one. In RNA, however, the sugar is ribose, which has an additional 2’ hydroxyl group. Not until 1952 did Todd show that 3’, 5’ bonds also join the ribonucleotides in RNA.
The double-helix model of DNA structure proposed by James Watson and Francis Crick in 1953 consists of two antiparallel polynucleotide chains joined by hydrogen
This absence of a 2’ hydroxyl group on deoxyribose meant that it was easier to synthesize oligodeoxyribonucleotides than it was to synthesize oligoribonucleotides. Even in this case, however, the functional groups attached to the purine and pyrimidine bases had to be protected (made chemically unreactive) in order to prevent side reactions with the hydroxyl groups of deoxyribose. To make a phosphodiester bond between the 3’ hydroxyl group of one nucleotide and the 5’ group of another, the 5’ hydroxyl group of the former and the 3’ group of the latter also had to be protected. In the 1956–1960 period, Smith and Khorana achieved the first synthesis of oligoribonucleotides (Smith and Khorana, 1959). They also developed a technique for synthesizing 3’, 5’ cyclic nucleotide monophosphates, which had been shown to be important intracellular regulators of glycogen metabolism.
In 1960 Smith married Helen Christie. The marriage produced three children: Tom, Ian, and Wendy. Smith and his wife separated in 1983 but never divorced.
Also in 1960 Khorana moved to the University of Wisconsin–Madison, taking Smith with him. Only a few months later, Smith returned to British Columbia to become head of the Chemistry Division at the Vancouver Technological Station of the Fisheries Board of Canada. There he initiated a research program on the pituitary hormones of salmon, but he also obtained external funding to continue his work on oligonucleotide synthesis. The Technological Station was situated on the campus of the University of British Columbia, and Smith formed links with members of the Department of Biochemistry, to which he was appointed as an associate professor in 1964. When Smith came into conflict with his superiors at the Fisheries Research Board concerning his research on nucleic acids, the head of the Department of Biochemistry successfully sponsored him for a research associate award from the Medical Research Council of Canada. In 1966 Smith took up his new position and moved into a laboratory in the Medical Sciences Building. He was promoted to the rank of professor in 1970 and awarded tenure the following year. Despite his falling-out with the Fisheries Board, Smith continued to work on the physiology and biochemistry of sexual maturation of salmon.
By the mid-1960s it was known that the DNA of genes was transcribed into messenger RNA, and the mRNA was then translated into protein. The genetic code—the relationship between the nucleotide sequences of DNA and RNA and the amino acid sequences of proteins—had also been elucidated. In recognition of the large part his synthetic oligoribonucleotides had played in solving the genetic code, Khorana shared the 1968 Nobel Prize in Physiology or Medicine. Smith realized that, because the nucleotide sequence of a gene is complementary to that of the corresponding mRNA, synthetic oligonucleotides might be used to bind to and thereby isolate specific mRNA molecules. He asked a PhD student, Caroline Astell, to determine the conditions under which synthetic deoxyribonucleotides could hybridize (form stable hydrogen-bonded duplex molecules).
In a series of studies published in the early 1970s, Astell and Smith showed that short oligonucleotides were capable of forming duplexes, the stabilities of which could be characterized by the temperature at which the two oligonucleotides dissociate, or “melt” (Astell and Smith, 1971). These melting temperatures were directly correlated with the length of the oligonucleotides and the amount of guanine and cytosine present. Two additional observations were critical for future developments. First, the melting temperatures of duplexes consisting of two oligodeoxyribonucleotides were much higher than those of duplexes containing one oligoribonucleotide and one oligodeoxyribonucleotide. From this Smith concluded that synthetic oligonucleotides would be more useful for isolating genes than for isolating RNAs. Second, stable duplexes could be formed between oligonucleotides containing a mismatched base, although the presence of such mismatches lowered the melting temperatures by approximately five degrees.
Chemical synthesis of oligonucleotides was time-consuming and inefficient. For this reason, Khorana had used a combination of chemical and enzymatic methods to produce polynucleotides large enough to function as templates for protein synthesis. The enzyme Khorana had used, DNA polymerase, was not suitable for making oligonucleotides of defined sequence. However, another enzyme, polynucleotide phosphorylase, had been shown to be capable of making oligoribonucleotides from ribonucleotide diphosphates. Studies involving Smith’s research associate, Shirley Gillam, established that, in the presence of the manganese ion, polynucleotide phosphorylase could be used to synthesize oligodeoxyribonucleotides (Gillam and Smith, 1972). As in Khorana’s studies, a combination of chemistry and biology produced the best results: chemical methods were used to make a short oligonucleotide which was then elongated by the enzyme. The elongation reactions were carried out in aqueous solution, and no protecting groups were required.
Synthetic oligonucleotides could only be used to isolate genes if the nucleotide sequence of the gene were known. In the early 1970s such information was extremely limited. Frederick Sanger of Cambridge University was developing a technique for the sequencing of genes, using the single-stranded DNA virus NX174 as a model. Sanger’s “plus-minus” method used enzymes to either extend or digest “primer” pieces of DNA. Following a visit to Smith’s laboratory in 1973, Sanger received some oligonucleotides, made using the method developed by Gillam, for use in the sequencing of the NX174 genome.
Motivated by a desire to learn DNA-sequencing methods, Smith spent the 1975–1976 academic year on sabbatical in Sanger’s laboratory. In what he later described as “the most fruitful and enjoyable scientific experience of my life” (Damer and Astell, 2004, p. 112), Smith was part of the team that determined the sequence of the 5,386 nucleotides in the NX174 genome (Brown and Smith, 1977). Another participant was Clyde Hutchison III, a virologist who was on sabbatical leave from the University of North Carolina. Hutchison had shown that if the bacterium Escherichia coli were infected with NX174 plus a fragment of NX174 DNA containing a mutation, a very small percentage of progeny viruses would “inherit” the mutation. As the replication of the virus involves a double-stranded DNA intermediate, this was interpreted to mean that occasionally the mutation-bearing DNA fragment would be incorporated into the new polynucleotide strand, and thereby into a progeny virus. In discussing this experiment, Smith and Hutchison came up with the idea of using synthetic oligonucleotides to introduce mutations into the DNA of NX174. The attraction of this approach was that it would allow any DNA sequence to be mutated.
Mutating DNA Back in Vancouver, Smith sent Gillam to Hutchison’s laboratory in Chapel Hill to learn how to handle E. coli and NX174. On her return, Smith attempted to utilize his group’s expertise in nucleic acid synthesis and hybridization to generate a specific mutation. The general idea was to synthesize an oligodeoxyribonucleotide that was complementary to part of a viral gene except for one mismatched base. Using the oligonucleotide as primer and single-stranded NX174 DNA as
template, DNA polymerase, he could then synthesize a polynucleotide complementary to the template strand except for the substitution of one base. If the resultant double-stranded DNA were used to infect E. coli, half the progeny viruses should carry the mutation. In order to screen for mutant viruses, Smith designed the synthetic oligonucleotide so that it was complementary to a known mutation, am3, which causes premature chain termination of a protein required for NX174 replication and is thus normally lethal. However, certain strains of E. coli were known to contain suppressor mutations allowing replication of viruses bearing this mutation. If Smith’s experiment worked perfectly, half the viruses produced would have acquired the am3 mutation, and thus would grow in the suppressor strain but not the wild-type strain of the bacterium (see Figure 1).
In fact, only a small percentage of progeny viruses was found to have the mutation. By treating the DNA to remove incomplete second strands, however, Smith and his colleagues were able to raise the efficiency of mutation to 15 percent. In an elegant additional experiment, they showed that synthetic oligonucleotides could also be used to revert mutant viruses back to wild type.
The paper describing these findings was published in the Journal of Biological Chemistry in September 1978— having first been rejected by Proceedings of the National Academy of Sciences and Cell. Clyde Hutchison was the first author, having won a coin toss to decide the priority of authorship (Hutchison et al., 1978).
The significance of site-specific mutagenesis—or site-directed mutagenesis, as it later became known—was not immediately recognized, perhaps because the initial method was only applicable to genes in certain viruses. In order to make it possible to mutate nonviral genes, Smith switched from NX174 to another bacteriophage, M13. NX174 is an icosahedral virus, with no room in the viral capsid for additional DNA, but M13 is filamentous, and a large amount of nonviral DNA can be incorporated into its filament. Therefore it was now possible to introduce a nonviral gene into M13 and then create a specific mutation using the appropriate oligonucleotide. Within five years of the first description of the technique, site-directed mutagenesis had been used to induce a variety of point mutations and deletions in genes isolated from organisms from bacteria to humans.
A paper published in Nature in 1982, describing a collaboration involving Smith and the enzymologist Alan Fersht, provided an early example of the power of site-directed mutagenesis in the genetic engineering of proteins (Winter et al., 1982). Fersht, Smith, and coworkers cloned the gene encoding the enzyme tyrosyl tRNA synthetase of the bacterium Bacillus stearothermophilus into M13, then used a synthetic oligonucleotide to create a mutation that changed a single amino acid at the active site of the enzyme. The mutant protein was then expressed, purified, and shown to have decreased catalytic activity. Automated methods of oligonucleotide synthesis, mutagenesis strategies based on the polymerase chain reaction, and other technological developments soon made site-directed mutagenesis a standard approach for correlating protein structure and function.
As noted above, Smith had concluded from Astell’s hybridization studies that synthetic oligonucleotides might be better suited to the isolation of genes than to the isolation of mRNAs. Contemporaneously with the development of site-directed mutagenesis, Smith collaborated with Benjamin Hall of the University of Washington to isolate the cytochrome c gene of yeast. Yeast was used because of the relatively small size of its genome; cytochrome c was chosen because partial sequence data were available. Using an oligodeoxyribonucleotide “13-mer” radiolabeled with phosphorus-32 to probe enzymatically generated fragments of yeast DNA, Hall, Smith, and coworkers were able to identify DNA fragments containing parts of the cytochrome c gene (Montgomery et al., 1978). In a subsequent study, the complete nucleotide sequence of the gene was determined. Continuation of the collaboration between Smith’s and Hall’s laboratories led to the elucidation of the three-dimensional structure of yeast cytochrome c and played an important role in the first identification of a factor controlling gene transcription in eukaryotes.
In 1981 Smith joined Hall and Earl Davie in founding the Seattle-based biotechnology company Zymos (later ZymoGenetics), with the aim of producing protein therapeutics such as recombinant human insulin. For several years Smith received a substantial consultancy fee from Zymos. He held stock in the company until 1988, when he sold his shares for a large sum that made him financially independent.
Honors Smith was increasingly recognized as a founder of molecular biology and biotechnology. He became a career investigator of the Medical Research Council of Canada in 1978 and received many awards for his research achievements. These included a Gairdner Foundation International Award and election as a fellow of the Royal Society, both in 1986. In his fifties, Smith started to devote more time to administrative activities and less to his own research program, although his laboratory remained highly productive. His administrative positions included director of the Centre for Molecular Genetics at UBC (1986–1987); director of the Biotechnology Laboratory, a center of excellence funded by the province of British Columbia (1987–1997); first scientific director of the federally funded Protein Engineering Network of Centres of Excellence (1990–1995); and interim scientific director of the Biomedical Research Centre at UBC (1991–1992).
The ultimate scientific accolade came in 1993, when Smith shared the Nobel Prize in Chemistry with Kary Mullis, inventor of the polymerase chain reaction. Already a wealthy man, Smith donated his share of the prize money to agencies supporting schizophrenia research, the training of women scientists, and the communication of science to the public. Shrewdly exploiting the publicity accompanying the Nobel award, he persuaded the governments of Canada and British Columbia to match his donations. The Canadian government also established the Michael Smith Award for Excellence, first awarded in 1994. In addition to the promotion of biomedical research, Smith used the prestige of his Nobel Prize to advance other causes in which he believed strongly, including peace and environmentalism.
The University of British Columbia recognized Smith’s Nobel award by granting him the honorary title of University Professor. In 1994 he was appointment Peter Wall Distinguished Professor of Biotechnology. By this time Smith was preparing to shed his administrative positions and was winding down his research program (he closed his laboratory in 1996). He used his freedom to go back to the bench, spending parts of 1996 and 1997 working in Maynard Olson’s laboratory at the University of Washington on the sequencing of human chromosome 7 and the genome of the bacterium Pseudomonas aeruginosa. Following this “sabbatical,” however, Smith was persuaded to take another administrative position, this time with the British Columbia Cancer Agency, which had launched a large-scale gene-sequencing program.
In 1998 Smith was diagnosed with myelodysplastic syndrome, an incurable form of cancer. Two years later this had developed into leukemia. After a brief period of hospitalization he died in October 2000.
Smith’s papers are held at the University of British Columbia Archives.
WORKS BY SMITH
With Har Gobind Khorana. “Specific Synthesis of the C5’-C3’ Inter-Ribonucleotide Linkage: the Synthesis of Uridylyl-(5’-3’)-Uridine.” Journal of the American Chemical Society 81 (1959): 2911–2912.
With Caroline Astell. “Thermal Elution of Complementary Sequences of Nucleic Acids from Cellulose Columns with Covalently Attached Oligonucleotides of Known Length and Sequence.” Journal of Biological Chemistry 246 (1971): 1944–1946.
With Shirley Gillam. “Enzymatic Synthesis of Deoxyribooligonucleotides of Defined Sequence.” Nature New Biology238 (1972): 233–234.
With Nigel L. Brown. “The Sequence of a Region of Bacteriophage NX174 DNA Coding for Parts of Genes A and B.” Journal of Molecular Biology 116 (1977): 1–30.
With Donna L. Montgomery, Benjamin D. Hall, and Shirley Gillam. “Identification and Isolation of the Yeast Cytochrome c Gene.” Cell 14 (1978): 673–680.
With Clyde A. Hutchison, Sandra Phillips, Marshall H. Edgell, et al. “Mutagenesis at a Specific Position in a DNA Sequence.” Journal of Biological Chemistry 253 (1978): 6551–6560.
With Greg Winter, Alan R. Fersht, Anthony J. Wilkinson, and Mark Zoller. “Redesigning Enzyme Structure by Site-Directed Mutagenesis: Tyrosyl tRNA Synthetase and ATP Binding.” Nature 299 (1982): 756–758.
“In Vitro Mutagenesis.” Annual Review of Genetics 19 (1985): 423–462.
Astell, Caroline. “Michael Smith.” Biographical Memoirs of Fellows of the Royal Society 47 (2001): 429–441.
Damer, Eric, and Caroline Astell. No Ordinary Mike: Michael Smith, Nobel Laureate. Vancouver, BC: Ronsdale Press, 2004.
Graeme K. Hunter