On this page
You learn the genetic code as a table: 64 codons, 20 amino acids, three stop signals. It is easy to forget that someone had to work out every entry, using chemistry, enzymes and crude extracts of bacteria. Har Gobind Khorana was one of the people who did it. This post covers his life, his method and how to predict the proteins his experiments produced, which is exactly the reasoning NEET questions on the code test.
Who was Har Gobind Khorana?
Khorana's Nobel biography says he was born in Raipur, a village in Punjab now in Pakistan. His exact date of birth is not known; documents give 9 January 1922. His father was a village agricultural taxation clerk, and the family was almost the only literate one in a village of about 100 people.
He earned an MSc at Punjab University in Lahore. In 1945 a Government of India fellowship took him to the University of Liverpool for a PhD. Postdoctoral work followed in Zurich with Vladimir Prelog and in Cambridge with G. W. Kenner and A. R. Todd. He then worked in Vancouver, moved to the University of Wisconsin in 1960 and to MIT in 1970. According to his Nobel facts page, he died on 9 November 2011.
What problem was he trying to solve?
In the 1950s it was established that DNA carries information that is passed through RNA to make proteins, and that a sequence of nucleotides corresponds to an amino acid. Marshall Nirenberg had found the first piece of the code. The Nobel facts page describes Khorana's contribution: he built different RNA chains with the help of enzymes, used them to make proteins, and the amino acid sequences of those proteins "solved the rest of the puzzle".
How did repeating sequences crack the code?
In his Nobel lecture, Khorana explained the method. His group chemically synthesised short pieces of DNA with repeating sequences. DNA polymerase copied and lengthened them into long, double-stranded DNA with the same repeat, and RNA polymerase then transcribed that into single-stranded RNA. These RNAs went into a cell-free protein-making system from E. coli.
The logic depended on the repeat:
A repeating dinucleotide, (AB)n, contains two triplets, ABA and BAB, in alternation. If the code is read in non-overlapping triplets, it should give a protein of two amino acids alternating.
A repeating trinucleotide, (ABC)n, can be read from three starting points: ABC, BCA or CAB. Each frame repeats one codon, so each should give a chain of a single amino acid, up to three such chains.
A repeating tetranucleotide should give a repeating four-amino-acid pattern, whichever frame is read.
He wrote that "All these predictions have been fully borne out experimentally without a single exception." From the results he concluded that DNA does specify the sequence of amino acids in proteins through RNA, and that the code is three-letter and non-overlapping. His group also synthesised all 64 possible trinucleotides for use in Nirenberg's binding test, which helped assign each codon.
Worked example: predicting the protein
Example 1: poly-UC (UCUCUC...). The triplets alternate: UCU, CUC, UCU, CUC. From the standard code, UCU is serine and CUC is leucine. So the product is a chain of alternating serine and leucine. This is the result Khorana showed for poly-UC, with serine and leucine incorporated in equal amounts.
Notice that this experiment alone cannot tell you which codon is serine and which is leucine. Khorana made the same point. The binding test with single trinucleotides settled it.
Example 2: poly-UUC (UUCUUCUUC...). Read from each starting point:
Frame 1: UUC UUC UUC... gives phenylalanine only.
Frame 2: UCU UCU UCU... gives serine only.
Frame 3: CUU CUU CUU... gives leucine only.
So you expect three separate chains: polyphenylalanine, polyserine and polyleucine. Khorana noted that repeating trinucleotide polymers "as a rule" gave three such chains.
Example 3: a stop codon in the repeat. Khorana reported that poly-UAG and poly-AUG gave only two amino acids. Read poly-AUG in its three frames: AUG, UGA, GAU. AUG is methionine and GAU is aspartic acid, but UGA is a stop codon, so that frame gives no chain.
Common mistakes to avoid:
Counting overlapping triplets. UCUC is not read as UCU, CUC, UCU starting at every letter. Each frame steps three letters at a time.
Forgetting the other frames. A repeating trinucleotide has three reading frames. Repeating dinucleotides give the same alternating protein from either start.
Ignoring stop codons. UAA, UAG and UGA end the chain. Always check each frame for them.
To practise more questions like these, our guide to the molecular basis of inheritance covers transcription, translation and the code. If you cannot see why the reading frame matters, the Fermi Concept Library covers every concept in biology with its own lesson, so you can revise that one idea and return.
What did he do after the code?
Khorana turned to making genes. In 1972 his group reported the total synthesis of the structural gene for an alanine transfer RNA from yeast90146-5) in the Journal of Molecular Biology. Building a gene from chemical pieces is one of the foundations of the biotechnology in your syllabus. Our post on the principles of biotechnology for NEET picks up from there.
What can a student learn from how Khorana worked?
Design the experiment so the answer is readable. Repeating sequences turned a messy problem into one with only a few possible outcomes.
Combine tools from different fields. Organic synthesis made the short pieces; enzymes made them long.
Credit the team. He began his lecture by calling his group's work "very much a group effort".
Try it now: predict the products of poly-AAG and say which amino acids you expect from each of its three reading frames.
