Author: Raleigh McElvery

Alex Rich had a long and fertile career at MIT working on the relationship between the molecular structure and the function of biological information molecules DNA and RNA. Rich is perhaps best known for the elucidation of the three-dimensional structure of a transfer RNA molecule, and for the discovery of an alternative form of DNA that exists in certain biological contexts, known as Z-DNA.

Less well-recognized is Rich’s contribution to the discovery of nucleic acid hybridization. Hybridization is the process by which single-stranded RNA or DNA molecules can find each other in solution by the exact matching of complementary base sequences. The rate of hybridization is limited only by the rate of diffusion of molecules in solution. Because of its remarkable speed and specificity, hybridization remains today as one of two fundamental methods for reading out the identity of RNA or DNA molecules in different contexts — the direct determination of the base sequence, with the other being carrying out the matching by computer.

Rich grew up in a working-class neighborhood of Springfield, Massachusetts. In high school, he helped support his family by working in the U.S. Armory machining grooves in rifle barrels. As a young man, Rich was smart, resourceful, and ambitious and he received a fellowship to Harvard College and later attended Harvard Medical School. At Harvard, Rich had the opportunity to work with Professor of Biological Chemistry John Edsall, who sparked an interest the physical chemistry of biological macromolecules that eventually led him away from medicine to postgraduate research with the visionary chemist and ebullient polymath Linus Pauling at Caltech.

Pauling discovered the alpha helix as a basic element of protein structure and by doing this invented the method of model building as a way of predicting the large-scale structural features of complex macromolecules from the chemical bonding structures of their constituent parts. In Pauling, Rich found a powerful role model who showed by example how far you could travel by grasping a good idea or deep insight.

When Rich joined the lab, Pauling was working on a structure for DNA and then was, to put it bluntly, scooped by James Watson and Francis Crick. Watson and Crick’s structure for DNA, based on astute model building and the X-ray diffraction data of Rosalind Franklin, was published in 1953. The key feature of their structure was the exact pairing of the bases between two strands of DNA that twist around each other in a double helix. The base-pairing rules — adenine pairs with thymine and guanine pairs with cytosine — are imposed by the geometric constraints on the paired bases as they are held together by hydrogen bonds in the central core of the helix. The double helix can accommodate a string of bases of any sequence and thus carry genetic information encoded in linear sequences of four characters. Moreover, the exact base-pairing between strands means that each strand carries the same information as the other, but in complementary form, and immediately suggested how the genetic information can be duplicated for cell division.

Big Bang and the coding problem

The emerging picture that the base sequence of DNA carried instructions to synthesize linear strings of protein out of a set of 20 amino acids led to a deeper puzzle: how could information encoded in the sequence of one kind of macromolecule be translated into the sequence of an entirely different kind of molecule? Although its direct involvement had not yet been shown, RNA was strongly suspected to have a central role in this process. One of the clues to the involvement of RNA was that RNA was most abundant in animal or plant tissues undergoing rapid growth and therefore extensive new protein synthesis. DNA and RNA are similar molecules and both are polymers of four nucleotide bases, but they differ in that DNA contains a hydrogen atom at the 2’ position on the ribose ring, whereas RNA contains a hydroxyl group at this position. The absence of a hydroxyl group at this position makes DNA more chemically stable and therefore more suitable to carry the permanent copy of genetic information. Also, DNA carries the base thymine instead of the chemically similar base uracil in RNA.

The brilliant theoretical cosmological physicist George Gamow, who was an early proponent of the Big Bang theory, saw that there was something worthy of interest in RNA and what soon became known as the “coding problem.” Gamow helped to focus thinking about this problem by posing the question of how a code written in four bases could be translated into 20 different amino acids. The introduction of the principles of information theory, first proposed in Claude Shannon’s 1948 paper “A Mathematical Theory of Communication,” immediately suggested that at least three bases would be required to carry enough information to specify 20 different amino acids. Gamow organized interested scientists in a group that called themselves the RNA Tie Club — so named for members’ necktie clips that bore the abbreviation of an amino acid. Rich was a member, as were Watson and Crick; and these members would share with one another ideas and insights before publication. Since each member of the club was assigned a different amino acid, membership never exceeded 20.

Rich was captivated by the connection, as so beautifully illustrated by the DNA model, between chemical structure and biological function and he was determined to make his mark in this new field as a structural biologist. He wondered if RNA could form a double helical base-paired structure and what role might this structure have in translating a DNA code into amino acids. With the help of Watson, who was at Caltech at that time, Rich set about analyzing different kinds of natural RNA samples, but none showed the characteristic diffraction pattern in X-ray analysis that Franklin had seen for double helical DNA.

Rich took a job at the NIH. There and on a sabbatical to Cambridge, England, he had success with various structural and modelling studies, including a structure for collagen, but he kept coming back to the question of whether a double-stranded RNA helix could form. One of the most precise analytical tools for nucleic acids such as RNA available at the time was to measure the base composition — that is, the relative proportion of guanine, cytosine, adenine and uracil. For a fully double-stranded molecule, base pairing rules would dictate that the amount of guanine should equal cytosine and the amount of adenine should equal that of uracil. The natural RNA samples that Rich was studying had very different base compositions, but did not follow the rules expected for a double-stranded structure. Eventually, Rich decided to force the issue by synthesizing his own RNA molecule that could form a fully base–paired double strand.

Rich and his colleague David Davies used the enzyme polynucleotide phosphorylase, which could polymerize into chains whatever activated nucleotide precursors were provided, to prepare two RNA chains designed to be able to base pair with each other. In one reaction, they prepared a long strand of RNA with only adinines (oligo-A) and in a separate reaction a long strand of RNA made up entirely of uracil (oligo-U). Hoping to see some amount of base pairing, Rich mixed the two preparations together and was amazed to see the entire contents become converted into RNA with the properties of double-stranded molecules. X-ray analysis confirmed that the two chains had had coiled around each other into a double helix. This experiment showed that an RNA-based double helix was possible, but the speed with which the double-stranded molecules formed was entirely unexpected. Based on physical chemistry of polymers, Rich had expected that some additional factors, such as enzymes, would be needed to neatly coil long disordered chains around each other. The effect of seeing this dramatic reorganization of molecules happen so efficiently might be the equivalent of seeing two tangled fishing lines that were thrown together spontaneously wrap themselves into a neat braid.

Such spontaneous base pairing between different nucleic acid chains is known as hybridization and is the fundamental underlying chemical process by which the information in DNA is translated into protein. Rich went on to show that hybridization between a DNA strand (oligo-dT) and an RNA strand (oligo-A) could occur to form a hybrid of RNA based paired with DNA. This molecule provided a structural basis for copying information from the gene sequence in DNA into a complementary single-stranded messenger RNA molecule. Moreover, base pairing between triplet codons on the messenger RNA and the anticodon loop of a transfer RNA carrying a specific amino acid is the basis by which the nucleotide code is translated into amino acid sequence.

Hybridization has become an enduring method in molecular biology and biotechnology research. Shortly after Rich carried out his RNA hybridization reaction, it was shown that the two strands of DNA could be melted apart at high temperature and then could come back together in a sequence specific manner if held at a somewhat lower annealing temperature. Before methods for direct sequencing of DNA became available, hybridization was the only method by which specific DNA or RNA sequences could be identified in a complex mixture.

Hybridization has become an enduring method in molecular biology and biotechnology research. Shortly after Rich carried out his RNA hybridization reaction, it was shown that the two strands of DNA could be melted apart at high temperature and then could come back together in a sequence-specific manner if held at a somewhat lower annealing temperature. Before methods for the direct sequencing of DNA became available, hybridization was the only method by which specific DNA or RNA sequences could be identified in a complex mixture.

Hybridization was crucial for the discovery of splicing of messenger RNA made by MIT Institute Professor Phil Sharp and was the basis for Professor Susumu Tonegawa’s demonstration of DNA rearrangements that underlie the formation of functional genes for antibodies. Even now, with extremely powerful methods for DNA sequencing, hybridization is still often used to examine the structure of chromosomes and to conduct comprehensive studies of gene expression based on microarrays. Finally, sequence-specific hybridization is at the heart of natural processes that have been harnessed for RNA interference of gene expression and CRISPR-based genome editing.

Rich himself wrote and spoke extensively about the early years of molecular biology in ways that reveal two important characteristics as a scientist. The first is that his deep admiration for mentors such as Pauling and colleagues such as Crick and Watson was the basis of an intellectual network that sustained Rich his entire career. By his account, a new discovery in the lab was invariably followed by a letter or a phone call to those that he admired to get their reactions. All biologists, no matter how great, struggle with the problem that it is difficult — if not impossible — when setting out on a new problem to predict whether it will reveal insights fundamental to all living things or merely lead to odd details produced as a byproduct of the tinkering of evolution. Rich was adept at vetting new ideas through his constellation of brilliant friends to guide him toward the fundamental.

The second, related characteristic is Rich’s gift for seeing how new concepts may play out in time — well into the future. In the spirit of the RNA Tie Club, Rich freely shared his imaginative speculation about where he saw the field going, adding these forward-thinking ideas to his review articles and sprinkling their seeds in the discussion sections of his research papers. Among his more prescient ideas was the prediction that hybridization to messenger RNA of a complementary regulatory RNA could play a part in gene regulation; this prediction anticipated the discovery of microRNA-based regulation by about 40 years. He also hypothesized in the early 1960s that early life forms could have a genetic system without DNA that was made up of only RNA. This may be the first articulation of the now widely accepted idea of an RNA world. As I knew Rich in his later years, he remained engaged in and stimulated by new ideas. It was not difficult when chatting with him in his office or going on a walk-and-talk with him to feel connected to and stimulated by the sweep of brilliant ideas that have propelled molecular biology along from the very beginning.

This story was originally published on the MIT School of Science.

The rise of the information age in the second half of the 20th century was spurred on by two related but distinct scientific and technological revolutions. The first, of course, was the digital revolution, which emerged with the development of the mathematics necessary for computation and data storage based entirely on a binary code. The second revolution came about from the discovery that information encoded in the molecular sequence of DNA carries the instructions for the working parts of a cell and thus is the blueprint of life. The field of molecular biology emerged as the study of how genetic information is transmitted from one generation to another and is read out to form functional cellular components and regulatory circuits.

The foundational science of molecular biology has led to methods for reading and writing biological information and to alter genomes by design. The capability to reprogram living organisms to do useful things forms the basis of the biotechnology industry.

No single institution has had a greater impact in accelerating the revolution in molecular biology and biotechnology than MIT. The origins of this revolution is woven deeply into the history of the Department of Biology.

MIT’s revolutionary foundation

In the late 1950s, MIT’s administration began a deliberate and concerted effort to recruit molecular biologists even before this nascent research area was recognized as a distinct field that would transform all of biology. The decision was made to hire faculty who were interested in studying biology by uncovering relationships between molecular structure and function and understanding the biochemical basis of genetic information and the transmission of genetic traits.

One such seminal hire was Alex Rich, the William Thompson Sedgwick Professor of Biophysics, who came to MIT in 1958. Rich contributed to the discovery of how single strands of DNA and RNA molecules can find and match complementary sequences. This process, called nucleic acid hybridization, remains one of the fundamental methods for reading out the identity of nucleic acid molecules. In addition to foundational research into hybridization, Rich also elucidated the three-dimensional structure of the transfer RNA molecule that functions in reading the genetic code.

A second enormously influential hire was Salvador Luria who moved from the University of Illinois to MIT in 1959. Luria was a leader in the study of bacteriophages — viruses that infect bacterial cells. Much in the same way that early quantum physicists used the hydrogen atom to establish a theory for quantum structure of atoms, the first molecular geneticists used the bacteriophage as a simple genetic system to reveal the rules for fundamental genetic processes such as replication, recombination, and mutation.

By the 1960s, the understanding of fundamental genetic mechanisms developed by Luria and others had merged with the work of structural biologists such as Rich to give the outline of how genetic information was stored, copied, and read.

The reading of genetic information takes place in two steps. In the first step, known as transcription, genetic information encoded in DNA is used as a template to make a copy in the form of a single-stranded messenger RNA. In the second step, the information contained within the messenger RNA is translated into a protein sequence at the site of protein synthesis — the ribosome. Transfer RNA molecules serve as the key adaptor molecules that allow translation of messenger RNA sequences into the amino acid sequences of proteins. Each transfer RNA carries a triplet of nucleotides that pairs with and thus “reads” a specific three-nucleotide sequence along the messenger RNA. At its other end, the transfer RNA carries a particular amino acid that is added in its place in the sequence of the elongating protein chain.

Khorana cracks the code

Before Har Gobind Khorana arrived in Cambridge, Massachusetts in 1970, he worked with the great nucleotide chemist, Alexander Todd at the University of Cambridge in the United Kingdom. Khorana was in the lab at the time that chemists were working out structure of the nucleotide building blocks of DNA and RNA. When Khorana started his own lab, first at University of British Columbia and then at University of Wisconsin, his work was devoted to using synthetic chemistry to make biologically important molecules and ever more complicated polynucleotide structures.

Khorana made one of the most consequential advances in molecular biology by using a hybrid approach that employed organic chemistry to synthesize short sequence of a few nucleotides followed by the use of a copying enzyme to generate long DNA molecules with many repeating copies of the short sequence. Khorana’s molecules with a repeating sequence were the keys to cracking the genetic code. A few years earlier, the complex process of translation was reconstituted in the test tube and was dependent on messenger RNA added from the outside. By using synthetic messenger RNAs to instruct the synthesis of proteins by the ribosome, Khorana’s group was able to work out rules for how specific sequences of three nucleotides in RNA are translated into the 20 possible amino acids. We now know that all forms of life use the same genetic code to read the information written in DNA. For his contributions to understanding the code, Khorana shared the 1968 Nobel Prize for Physiology or Medicine.

Synthesizing genes

As work on the code was nearing completion, Khorana began thinking about how to synthesize long polynucleotide molecules of even greater complexity. He had his eye on what could be considered a moonshot challenge in nucleic acid synthesis: to synthesize a functional gene.

Before an artificial gene could be synthesized, it was necessary, of course, to know the DNA sequence of the desired gene. In the mid-1960s, the ability to directly determine the DNA sequence of a protein-coding gene was still about a decade away; however, the DNA sequence of an RNA-coding gene could be deduced directly from the RNA sequence. The first complete sequence of a natural gene-encoded RNA molecule — the transfer RNA for the amino acid alanine — was determined by Robert Holley in 1965. In that year, Khorana began to organize his lab to synthesize the double-stranded DNA that would code for alanine transfer RNA. Although Khorana knew that this monumental task would require a combined and concerted effort of perhaps a decade of work, he expressed utter clarity and confidence in the purpose and significance of this endeavor.

In a review letter for the Biochemical Journal in 1968, he wrote: We would like to know, for example, what the initiation and termination signals for RNA polymerase are, what kind of sequences are recognized by repressors, by host modification and host restrictive enzymes, and by enzymes involved in genetic recombination, and so on. For these studies, ultimately what is required is the ability to synthesize long chains of DNA with specific non-repeating sequences. With this should come the ability to ‘manipulate’ DNA for different types of studies.

This description pretty well summarizes the work of a major segment of molecular biology for the next 50 years.

In theory, the DNA for alanine transfer RNA could be formed by synthesizing each complementary strand separately and then using hybridization to form a complete double-stranded helix. This approach would require synthesis of DNA strands that were 77 nucleotides long; however, at the time the upper limit for synthesis, even in Khorana’s laboratory, was about 20. The plan as originally conceived was to take advantage of the ability of DNA polymerase to synthesize DNA from a template. The idea was to synthesize oligo-nucleotides that partly overlapped and then to use DNA polymerase to complete a fully double-stranded DNA molecule. Khorana’s team started the synthesis of the gene for alanine transfer RNA in this way and showed that basic strategy of using chemical synthesis followed by synthesis by polymerase would work. But when the DNA ligase enzyme was discovered, it became more practical to chemically synthesize many short overlapping segments and stitch them together with ligase. In this manner, the synthesis of alanine transfer RNA gene was completed in 1970.

The first synthetic gene was in itself a monumental landmark in the progression of molecular biology; but like any successful moonshot, the technological innovations developed along the way may have had the furthest-reaching impact.

Knock-on effects

Marvin Caruthers joined Khorana part of the team synthesizing the alanine transfer RNA in 1966 and then came with him to MIT. Caruthers then went to the University of Colorado at Boulder, where he began his own research program developing methods for reliable automated synthesis of short DNA molecules, or oligonucleotides. He decided to carry out nucleotide synthesis on a solid support, which would greatly simplify and speed up the separation of the growing oligo-nucleotide chain away from precursor molecules as the process stepped through the reaction cycle for the addition of each base in the sequence.

A second key innovation was Caruther’s development of nucleotide precursors that could be stored for long periods and then readily activated immediately before use. The so called “phosphoramidite method” for DNA synthesis was automated and its use enables scientists who are not expert organic chemists to synthesize their own oligonucleotides. The ready availability of oligonucleotide primers has driven the expansion of methods for reading DNA by sequencing and the copying and modification of DNA sequences at will. These technologies are analogous to the fundamental output and input devices of a digital computer but for the manipulation of biological information encoded in DNA.

The development of the by polymerase chain reaction (PCR) is another key technological advance that stemmed from Khorana’s work. PCR employs the same basic elements proposed by Khorana for the synthesis of the alanine transfer RNA gene; hybridization of synthetic oligonucleotides to a target DNA followed by synthesis with DNA polymerase to produce double-stranded DNA of defined sequence. The key innovation as proposed by Kary Mullis when he came up with the idea for PCR was to use the same synthetic oligonucleotides to conduct many cycles of hybridization and synthesis. Because of the doubling that results from each round of replication, 20 cycles would give a million-fold amplification allowing a specific sequence to be produced from an extremely complex mixture such as a whole genome.

Twelve years before the invention of PCR, Khorana’s group showed that oligonucleotides defining the ends of the completed transfer RNA gene segment could be used to carry out rounds of hybridization and DNA synthesis with polymerase to make more of the desired DNA product without any additional labor in chemical synthesis of DNA. This raises the question of whether Khorana, who was a visionary, foresaw the possible application of his method for the amplification of sequences from whole genomes. It is worth pointing out that at the time Khorana’s group was contemplating enzymatic amplification, their synthetic gene was one of the only DNA sequences that was known and therefore a basic ingredient of the PCR method — knowledge of enough of an interesting target sequence to design the oligonucleotide primers for its amplification — was not available to them. Years later, when PCR patents were under litigation, the question of prior art arose; but Khorana refrained from comment, having moved on to the study of the light-sensing protein rhodopsin.

Visions for the next revolution

At a memorial service for Khorana held at MIT in 2012, many stories were told about his intellectual independence and visionary leadership in basic research that had far-reaching implications.

As the synthesis of alanine transfer RNA gene was well underway, Khorana initiated a project reaching for an even bigger prize — a synthetic gene that could be shown to carry out its biological function in the context of a living cell. The candidate, known as a suppressor transfer RNA, was a recently sequenced transfer RNA that had the ability to read past a stop mutation introduced in the middle of a gene, thereby suppressing the effect of the mutation and allowing ribosomes to read the RNA and produce the protein. The idea that Khorana laid out for the team was to synthesize the suppressor transfer RNA and then introduce the synthetic gene into a suitable bacterial host designed to test its ability to suppress a stop mutation.

At that time, now standard methods for gene cloning and expression did not exist. As the planning moved forward, the team synthesizing the suppressor transfer RNA began to envision more and more elaborate schemes to get a functional suppressor transfer RNA gene into cells. As Caruthers related the story, Khorana listened quietly to the brainstorming for a bit and then said, “Let’s first synthesize the gene. By that time, we will know how to express it.” Khorana was right; and by the time the synthetic suppressor gene was complete, methods were available for introducing the gene into cells.

Like the great explorers Frances Drake and Ernest Shackleton who were my heroes growing up, Khorana had the vision and leadership to convince a team to follow him to an unknown place, and he had the supreme confidence that he would know what to do once he got there.

Transformational scientific and technological revolutions, like those initiated by Khorana, Luria, and Rich, are of keen interest because they help us understand the sparks of genius and originality that we should be looking for when we hire new faculty and illustrate the kinds of research projects in our institutions and companies that might lead to fundamental advances in preparation for the next scientific revolution.

Chris Kaiser is the Amgen Inc. Professor of Biology and the former head of the MIT Department of Biology.