Finding the gene and getting the DNA out

Step one: choose the gene. A candidate gene is chosen because of its value in medicine, farming or research: the gene for insulin (for diabetes), genes for salt tolerance or disease resistance in crops, genes for milk proteins or blood-clotting factors. Biochemical and physiological studies point to it.

Then it has to be found. A typical gene is only 3000-3500 base pairs long, hidden in a human genome of 3.2 billion: harder than finding a needle in a haystack. Two ways help:

• From the protein's amino acid sequence, work backwards to the DNA sequence that codes for it.
• Take the mRNA made in the tissue where the gene is active, and copy it into DNA with the enzyme reverse transcriptase. This copy is cDNA (complementary DNA).

Step two: get the DNA out. This is harder than it sounds. Nucleic acids are present in small amounts, are very long and fragile, are easily damaged by chemicals, and the cell's own nucleases (enzymes that destroy them) are released at the same time. Extraction has four stages: break the cells, protect the nucleic acids, separate them from everything else, and precipitate them.

• Break the cells (lysis). Bacteria: the enzyme lysozyme and the detergent SDS. Plants: grinding, and the detergent CTAB (with PVP to remove plant phenols). Animal cells: a detergent. The buffer is mildly alkaline and contains EDTA, which grabs Mg²⁺ and Mn²⁺ ions; nucleases need these ions, so they cannot work.
• Remove proteins. SDS gives proteins a negative charge so they let go of the DNA; high salt completes this. A mixture of chloroform and isoamyl alcohol is added and the tube spun: proteins collect in the lower, organic layer, and nucleic acids stay in the upper, watery layer.
• Precipitate. Adding ethanol makes the DNA come out of solution as white threads. Ribonuclease A is added to destroy any RNA.

RNA is harder to isolate, because enzymes that destroy it, RNases, are everywhere, even on our skin. A mixture of GITC, phenol and chloroform keeps RNA in the watery layer, while DNA and proteins go to the boundary and the organic layer; isopropanol then precipitates the RNA.

To get plasmid DNA, it must be separated from the much larger chromosomal DNA: for example, by boiling, or by breaking cells with SDS and NaOH and then removing the chromosomal DNA with potassium salts.