Biotechnology & Genetic Engineering: Chapter 2 — Cutting, joining, and copying DNA

2.1 Cutting and joining #

2.1.1 Restriction–modification is a bacterial system #

Many bacteria cut incoming DNA at particular sequences and methylate the same sequences in their own genome. Phage grown on the same strain is protected; phage grown elsewhere often is not. Arber, Smith, and Nathans shared a Nobel prize for that genetics and for the enzymes. The system is a defence. Cloning uses the accident that one class of the enzymes cuts a predictable bond.

Type IIS geometry, and later overlap methods, still depend on compatible ends and sealing. These notes do not specify those assemblies as procedures.

2.1.2 Sticky ends, blunt ends, and how often a site appears #

Sites are often palindromes of four, six, or eight base pairs. In random DNA those are expected about every 256 bp, 4 kb, and 65 kb. Real genomes are not random, so the frequency moves with base composition.

A staggered cut leaves a sticky 5′ or 3′ overhang. Matching overhangs base-pair before ligation, which makes the join efficient and selective. Blunt ends can join any other blunt end, less efficiently. BamHI and BglII leave the same overhang; the hybrid junction is cut by neither. Isoschizomers recognise the same site; neoschizomers cut it differently. Dam methylation can hide a GATC site. Star activity, cutting at related sites under poor conditions, creates extra bands. A gel tests the map.

2.1.3 DNA ligase #

Ligase joins a 3′ hydroxyl to a 5′ phosphate. Without the phosphate it has nothing to seal. E. coli ligase uses NAD+ and is poor on blunt ends. T4 ligase uses ATP and seals both, which is why it is the cloning enzyme. The adenylate chemistry belongs to Biochemistry. Both strands must be closed or the cell will not return a stable plasmid.

Ligation is promiscuous: vector to itself, insert to insert, and the wrong orientation of a blunt or palindromic end. Selection exists for that reason. Removing vector phosphates so the backbone cannot recircularise alone is logic, not a protocol given here.

2.2 Copying a sequence and writing a short one #

2.2.1 PCR as a principle #

PCR copies the DNA between two primers, short oligonucleotides that base-pair and provide a 3′ hydroxyl. The cycle denatures the template near 95°C, anneals primers near the temperature where that duplex is stable, and extends, commonly near 72°C for Taq polymerase from Thermus aquaticus. Products become templates, so the fragment grows exponentially until something runs out.

Early PCR used a heat-killed polymerase and fresh enzyme each cycle. Taq survives 95°C, which is what made the method routine. Mullis shared the 1993 Nobel prize for it; Smith shared it for site-directed mutagenesis. Taq lacks proofreading, errs about once in 10^4 to 10^5 bases, and often leaves a 3′ A. Proofreading archaeal polymerases err less, often by about tenfold to a hundredfold, and leave blunt ends. Amplified does not mean faithful.

PCR copies a segment whose ends match the primers, if the template is present and accessible. It cannot copy a sequence with no primer match, or RNA until reverse transcriptase has acted. Promiscuous primers amplify whatever they actually primed, and short unintended products outrun long ones. A genome is not one PCR product. A negative result can be inhibition. Exponential gain means a few contaminant molecules can become the whole signal. No primer sequences are given here.

2.2.2 Reverse transcriptase and cDNA #

Reverse transcriptase copies RNA into DNA. A cDNA is that copy, usually of mRNA. Oligo(dT) favours polyadenylated mRNA; random primers copy more of the RNA population; a gene-specific primer copies one transcript. Second-strand synthesis replaces the RNA so the product can be cloned.

🔬 Key Illustrated Scientific Figure

Full-Screen Zoom

Click to enlarge on this page

cDNA is not the chromosomal gene. Introns are already gone, which is why the coding sequence can be expressed in E. coli. Promoters are mostly absent. The 5′ end is often incomplete. Abundance follows the transcript in that sample, so a silent gene is missing and a loud mRNA dominates. Chapter 3 uses that bias.

2.2.3 Chemical synthesis, and why length is the limit #

Phosphoramidite synthesis builds DNA 3′ to 5′ on a solid support, the opposite of polymerase. Each cycle deprotects, couples, caps failures, and oxidises. Full-length yield is about the per-step yield raised to the number of couplings. At about 99% per step a 20-mer is mostly full length; a 100-mer made as one chain is rich in deletions. Acidic deprotection also damages purines slowly.

Primers, probes, and single-codon changes are routine. A kilobase gene is assembled from short oligonucleotides or copied as cDNA. Synthesis writes what you specify. It does not judge whether that sequence should be made. Chapter 10 puts designs aimed at a more dangerous organism outside ordinary work.

2.3 Key terms, summary, and self-check #

Key terms. Restriction–modification; type II enzyme; sticky end; blunt end; compatible ends; isoschizomer; star activity; DNA ligase; PCR; Taq polymerase; reverse transcriptase; cDNA; phosphoramidite synthesis; coupling efficiency.

Summary. Type II enzymes cut a defined site and leave ends ligase can seal. Site frequency falls by a factor of about sixteen for every two extra bases, and real genomes depart from that. PCR amplifies whatever two primers actually bracket, exponentially, with error and contamination as limits. cDNA is a spliced, abundance-biased copy of RNA. Chemical synthesis is reliable for short strands and fails as one long chain because a small coupling loss compounds.

Self-check.

1. Why is a type I enzyme a poor cloning cutter?

2. What does a sticky end add, and how can a BamHI–BglII join destroy both parent sites?

3. Which groups does ligase join, and why is T4 ligase used on blunt ends?

4. Give three things PCR cannot do even if the cycler is correct.

5. Why is a 20-mer a routine synthetic product and a 2 kb gene not one chain?

—

—

—