Biotechnology and Its Issues

Restriction enzymes (restriction endonucleases) are bacterial enzymes that cut DNA at specific recognition sites, usually 4–8 base pairs long. In nature they defend bacteria by chopping up the DNA of invading viruses; the bacterium protects its own DNA by adding methyl groups to its own recognition sites. Most recognition sites are palindromes: the sequence reads the same 5'→3' on both strands. EcoRI (from E. coli) recognizes 5'-GAATTC-3' and cuts between the G and the A on each strand. Because the two cuts are offset, each piece is left with a short single-stranded overhang, 5'-AATT, called a sticky end. Sticky ends made by the same enzyme are complementary, so any two pieces cut with that enzyme can base-pair with each other. Other enzymes, such as SmaI (which cuts CCC|GGG), cut straight across both strands and leave blunt ends with no overhang; blunt ends can be joined, but less efficiently. DNA ligase then forms phosphodiester bonds to seal the sugar-phosphate backbones, permanently joining the pieces. In a random sequence, any particular 6-base site appears on average once every $4^6=4096$ base pairs, so a six-base cutter such as EcoRI makes fragments a few thousand base pairs long.

To make recombinant DNA (DNA that combines sequences from different sources), the gene of interest and a vector are cut with the same restriction enzyme. A common vector is a plasmid, a small circular DNA molecule that bacteria copy independently of their chromosome. A cloning plasmid has an origin of replication, a site for the restriction enzyme, and a selectable marker such as a gene for resistance to the antibiotic ampicillin. The cut gene and cut plasmid are mixed so their sticky ends pair, and ligase seals them. Bacteria are then transformed: treated with calcium chloride and a brief heat shock (or an electric pulse) so that some take up plasmids. Only a small fraction succeed, so the bacteria are spread on agar containing ampicillin. Cells without a plasmid die, and only transformed cells grow into colonies. For human insulin, the gene used is a copy made from the insulin mRNA with the enzyme reverse transcriptase, called complementary DNA (cDNA). This matters because bacteria cannot splice out introns, and cDNA made from mature mRNA has none. The transformed bacteria are grown in large fermentation tanks, and the insulin is purified.

Gel electrophoresis separates DNA fragments by size. Samples are loaded into wells at one end of a slab of agarose gel, which sits in a buffer between two electrodes. DNA is negatively charged because of its phosphate groups, so when the current is on, every fragment moves toward the positive electrode; the wells are placed at the negative end. The gel is a tangled mesh of fibres: short fragments slip through easily and travel farther, while long fragments are held back. Every DNA molecule has the same charge per unit length, so size is the only thing that decides how far a fragment moves. A DNA ladder, a mixture of fragments of known sizes, runs in one lane as a ruler. Afterward the gel is stained with a dye that glows under UV or blue light. Each visible band is millions of copies of fragments of one size.

The polymerase chain reaction (PCR), invented by Kary Mullis in 1983, copies a chosen stretch of DNA billions of times in a few hours. The tube holds the template DNA, two short DNA primers that flank the target sequence, all four nucleotides (dNTPs), and Taq polymerase, a heat-stable DNA polymerase from Thermus aquaticus, a bacterium from hot springs. A thermal cycler repeats three steps: denaturation at about 95 °C separates the strands; annealing at about 55–65 °C lets the primers pair with their target sequences; extension at about 72 °C lets Taq build new strands from the primers. Each cycle doubles the number of copies, so after $n$ cycles $N=N_0\times 2^n$; 30 cycles turn one molecule into about $2^{30}\approx 1.07\times 10^9$. Sanger sequencing (1977) reads the order of bases. A polymerase copies the template in the presence of a small amount of dideoxynucleotides (ddNTPs), which lack the 3'-OH needed to add the next nucleotide, so a strand stops wherever one is added. The result is a set of fragments ending at every position, each end tagged with a colour for its base; sorted by size, they spell out the sequence. DNA fingerprinting compares short tandem repeats (STRs), places where a 2–6 base sequence is repeated back to back a number of times that varies between people. PCR copies about 20 STR loci, and the combined pattern is so variable that the chance of two unrelated people matching is far below one in a billion. Canada's National DNA Data Bank, run by the RCMP since 2000, stores STR profiles, and DNA evidence helped overturn the wrongful convictions of Guy Paul Morin (1995) and David Milgaard (1997).