№ 159 · biology

How a genome gets read

A cell reads its genome by copying genes into mobile RNA messages, which ribosomes then turn into protein. The surprise is that in eukaryotes the copy is not a plain duplicate: the raw transcript is capped, tailed and spliced. A laboratory sequencer also reads DNA, but in a different sense, covered under “Reading genomes at scale”.

The central dogma

Fowler, Roush and Wise describe the flow of information as DNA to mRNA to protein. A gene specifies the sequence of an mRNA, and that mRNA specifies a protein. Copying DNA into mRNA is fairly simple: one nucleotide is added for each complementary nucleotide on the DNA strand being read. Building the protein is more involved, because each amino acid matches a group of three mRNA nucleotides.

Transcription, in three stages

The double helix partly unwinds at a transcription bubble, and proteins bind a promoter, a DNA sequence that sits upstream of the gene in most cases. The promoter's sequence decides how often that gene is transcribed. During elongation, RNA polymerase moves along one DNA strand, the template strand, adding nucleotides by base pairing. The product matches the other strand except that RNA carries uracil where DNA has thymine. The DNA unwinds ahead of the enzyme and rewinds behind it. In prokaryotes, termination signals come in two kinds; both are repeated sequences in the template that make the polymerase stall, leave the DNA and release the transcript.

Predict first

In a eukaryotic cell, can a ribosome start translating an mRNA while RNA polymerase is still copying it?

Check your answer

No. In prokaryotes the two can run at once, but in eukaryotes the nuclear membrane prevents simultaneous transcription and translation.

Editing before export

Fowler, Roush and Wise note that eukaryotic transcripts are processed before leaving the nucleus. A cap is added to the 5' end while the transcript is still being made, and proteins coat it to stop it degrading. After elongation, roughly 200 adenine residues are added at the 3' end as a poly-A tail. Genes also contain exons, which are expressed, and introns, which lie between them. Splicing removes the introns and joins the exons; an error of a single nucleotide shifts the joined sequence and spoils the protein. The same authors add that these extra steps leave the message far more stable: eukaryotic mRNA lasts several hours, against five seconds at most for a typical prokaryotic mRNA.

Translation and the code

Translation needs mRNA, ribosomes, tRNAs and enzymatic factors. A ribosome has a small subunit that binds the mRNA and a large subunit that binds tRNAs, which carry amino acids and pair with codons. Each amino acid is defined by a triplet codon, and with four letters there are 64 possible triplets, so one amino acid can be encoded by more than one. Three codons stop synthesis; AUG both codes for methionine and marks the start. Fowler, Roush and Wise report that E. coli adds each amino acid in 0.05 seconds, so a 200-amino-acid chain could be built in 10 seconds.

Interactive Drag the slider to move the ribosome along a short illustrative mRNA and watch the growing chain.

An illustrative mRNA of eight codons, with a ribosome box on the codon the slider selects and the chain built so far drawn along the top. The codons are a made-up sequence, not a real gene; the amino acids are assigned using the genetic-code table in Fowler, Roush and Wise's Figure 9.20. The timing readout uses the 0.05 s per amino acid that Fowler, Roush and Wise (2013) give for E. coli, applied here to this illustrative chain.

Reading genomes at scale

Everything above is the cell reading its own genes to make proteins. Sequencing is a different kind of reading: a laboratory instrument works out the sequence of DNA fragments so that people can study it. Satam and colleagues review the generations as of 2023. Sanger's chain-termination method, using dideoxynucleotides, gave reads at most a few hundred nucleotides long, and the ABI 370 automated it in 1987. Second-generation platforms sequence many fragments in parallel. Third-generation methods give long reads: the review puts long-read output at up to 25–30 kb against roughly 600–700 bp for short reads. The authors also judge that, as of 2023, next-generation sequencing (NGS) in its current form cannot stand in for standard point-of-care tests such as PCR when infectious pathogens must be diagnosed quickly.

In short

DNA is copied into RNA starting at a promoter, edited in eukaryotes, then read three nucleotides at a time by ribosomes. Sequencing is a separate, laboratory kind of reading: instruments work out the sequence of many DNA fragments, and Satam and colleagues describe long-read methods, as of 2023, that read much longer pieces than short-read ones.

Where this comes from

  1. Concepts of Biology, 9.3 Transcription (OpenStax) linked only, not reproduced
    Fowler, Roush, Wise · 2013
    openstax.org/books/concepts-biology/pages/9-3-transcription
  2. Concepts of Biology, 9.4 Translation (OpenStax) linked only, not reproduced
    Fowler, Roush, Wise · 2013
    openstax.org/books/concepts-biology/pages/9-4-translation
  3. Next-Generation Sequencing Technology: Current Trends and Advancements (Biology 12:997) linked only, not reproduced
    Satam, Joshi, Mangrolia, Waghoo, Zaidi, Rawool, Thakare, Banday, Mishra, Das, Malonia · 2023
    pmc.ncbi.nlm.nih.gov/articles/PMC10376292/