Recombinant DNA Technology and Molecular Tools — Enzymes, Vectors, Cloning, PCR, Blotting and Genome Editing
2. Cloning vectors — plasmids, phage, cosmids, Ti, BAC, YAC, and expression vectors
A cloning vector must have an origin of replication, one or more selectable markers, and a multiple cloning site (polylinker) with unique restriction sites. Plasmids (pUC, pBR322) carry inserts up to roughly 10 kb and are the everyday vector; pUC vectors add blue–white screening through the lacZ-alpha fragment, where an insert in the polylinker disrupts beta-galactosidase so the colony stays white on X-gal while empty vector turns blue. Bacteriophage lambda vectors take about 9–23 kb and package DNA efficiently in vitro, which is why they suit libraries; cosmids combine a plasmid origin with lambda cos sites to package up to about 45 kb. Larger still are BACs (bacterial artificial chromosomes, based on the F plasmid, ~100–300 kb) and YACs (yeast artificial chromosomes, which have a centromere, two telomeres and an origin and carry hundreds of kb to over a megabase). The larger the insert the vector takes, the fewer clones a whole genome needs, but the harder the DNA is to handle.
An expression vector does more than carry a gene — it makes the host transcribe and translate it. It adds a promoter (strong and often regulatable, such as the lac, tac or T7 promoters in E. coli), a ribosome-binding site at the right spacing from the start codon, and a terminator; a eukaryotic expression vector adds a polyadenylation signal and often an enhancer. Useful extras are an affinity tag (a His₆ tag for nickel-affinity purification, GST, or an epitope tag), a signal/secretion sequence to export the protein, and a protease site to cleave the tag off afterwards. The Ti plasmid of Agrobacterium (Section 5) is the vector for making transgenic plants; the choice of promoter and host is what separates a vector that merely clones from one that produces protein.
3. Libraries, gene isolation and cloning; transposons, gene targeting and recombination-based cloning
A genomic library is a collection of clones covering an organism’s entire genome; a cDNA library is made by reverse-transcribing the mRNA of a cell, so it holds only the genes expressed there and, in eukaryotes, only the exons (no introns) — which is why a human gene for expression in bacteria is cloned from cDNA, since bacteria cannot splice. The number of clones needed to cover a genome with a given probability follows the Clarke–Carbon formula N = ln(1 − P)/ln(1 − f), where f is the fraction of the genome in one insert and P the desired probability of coverage. A gene is found in a library by screening — colony or plaque hybridisation with a labelled probe, antibody screening of an expression library, or PCR with gene-specific primers.
Strategies for producing recombinant proteins combine the pieces above: clone the coding sequence (cDNA for a eukaryotic protein) into an expression vector with a suitable promoter and tag, transform the right host (E. coli, yeast or mammalian cells — Section 5), induce expression, and purify (Section 5 downstream processing). Codon optimisation, choice of a soluble-expression strategy and secretion all raise yield. Transposons — mobile DNA elements that move by a transposase — are used as tools for insertional mutagenesis and gene tagging, interrupting a gene and marking its location. Gene targeting replaces or disrupts a chosen gene by homologous recombination (Section 5, knock-outs). Recombination-based cloning — the Gateway system (using phage lambda att sites and its integrase/excisionase) and Cre–lox and Gibson assembly (an exonuclease, polymerase and ligase joining overlapping fragments in one reaction) — moves or assembles DNA without relying on unique restriction sites, which is what makes modern, restriction-site-independent construction possible.
4. PCR, nucleic acid labelling, and Sanger and next-generation sequencing
The polymerase chain reaction (PCR) amplifies a defined DNA region exponentially through repeated cycles of three steps: denaturation (~94–95 °C, strands separate), annealing (~50–65 °C, two primers bind the ends of the target on opposite strands) and extension (~72 °C, a thermostable polymerase such as Taq copies each strand 5-prime to 3-prime). Because each cycle copies both new and old strands, the target doubles each cycle: after n cycles the amount is multiplied by 2ⁿ in the ideal case (the specific product bounded by the two primers accumulates exponentially after the first few cycles). Annealing temperature is set a few degrees below the primers’ melting temperature; too low gives non-specific products. Variants: RT-PCR (reverse-transcribe RNA first, to amplify from mRNA), qPCR / real-time PCR (a fluorescent dye or probe tracks product each cycle, and the threshold cycle Ct is lower the more template was present, so Ct quantifies starting amount), and nested, multiplex and colony PCR.
Nucleic acids are labelled for use as probes — radioactively (³²P by nick translation, random priming or end-labelling with polynucleotide kinase) or non-radioactively (biotin or digoxigenin detected by enzyme-linked antibodies, or a direct fluorophore). Sanger (dideoxy) sequencing copies the template with DNA polymerase in the presence of a small fraction of chain-terminating dideoxynucleotides (ddNTPs), which lack the 3-prime hydroxyl, so synthesis stops wherever one is added; the mixture of fragments of every length is separated by size (capillary electrophoresis), and in modern dye-terminator sequencing each ddNTP carries a different fluorophore, so the sequence is read off as a ladder of coloured peaks. Next-generation sequencing (NGS) sequences millions of fragments in parallel — Illumina uses bridge amplification and reversible dye-terminator sequencing-by-synthesis to read short reads with very high throughput; long-read platforms (nanopore, single-molecule real-time) read far longer fragments at the cost of higher per-base error. NGS made whole genomes, transcriptomes (RNA-seq) and metagenomes routine.
5. Blotting, in-situ hybridization and DNA fingerprinting (RAPD, RFLP)
| Blot | Target | Separated then detected by |
|---|---|---|
| Southern | DNA | Gel electrophoresis, transfer to membrane, hybridisation with a labelled DNA probe |
| Northern | RNA | Same, with a probe, to measure a transcript’s size and amount |
| Western | Protein | SDS-PAGE, transfer, detection with an antibody (see Chapter 2) |
Southern blotting (named for Edwin Southern; the others are named by analogy) separates restriction-digested DNA on a gel, transfers it to a membrane and detects a chosen sequence with a labelled probe — it answers "is this sequence present, and in what size fragment?". Northern blotting does the same for RNA and so measures a gene’s expression. In-situ hybridisation applies a labelled probe to DNA or RNA inside fixed cells or tissue, so a transcript or a chromosomal locus is located in place; FISH (fluorescence in-situ hybridisation) uses fluorescent probes on chromosomes to map a gene or detect copy-number changes.
DNA fingerprinting distinguishes individuals by variation in repeated sequences. RFLP (restriction fragment length polymorphism) reads a mutation that creates or destroys a restriction site, or a length difference in a repeat, as a change in the size of the fragments a restriction enzyme produces, revealed by a Southern blot; it needs good-quality DNA and is laborious but highly reproducible and codominant. RAPD (random amplified polymorphic DNA) uses PCR with single short primers of arbitrary sequence at low annealing temperature, which prime wherever they find an approximate match, so the pattern of bands reflects polymorphisms across the genome; it needs no sequence knowledge and little DNA but is dominant and sensitive to conditions, so less reproducible between labs. Modern profiling has largely moved to STR (microsatellite) markers read by capillary electrophoresis.
6. Site-directed mutagenesis, CRISPR–Cas, biosensors, interaction tools and omics approaches
Site-directed mutagenesis introduces a defined change — a point mutation, insertion or deletion — at a chosen position, to probe a protein’s structure–function. The classic method uses an oligonucleotide primer carrying the desired change, which primes synthesis of a new strand on the template so the mutation is copied in; modern kits amplify the whole plasmid with mutagenic primers and then digest the methylated parental template with DpnI, which cuts only methylated DNA, leaving the mutant. CRISPR–Cas9 is programmable genome editing: a guide RNA base-pairs with a 20-nt target that must sit next to a short PAM sequence, and the Cas9 nuclease makes a double-strand break there. The cell repairs it by non-homologous end joining (NHEJ), which is error-prone and so knocks a gene out by small indels, or, if a donor template is supplied, by homology-directed repair (HDR), which knocks a precise change in. Catalytically dead Cas9 (dCas9) fused to effectors instead represses, activates or edits bases without cutting.
A biosensor couples a biological recognition element (enzyme, antibody, nucleic acid, cell or receptor) to a transducer that turns the binding or reaction into a measurable electrochemical, optical, piezoelectric or thermal signal. The glucose biosensor — glucose oxidase on an electrode measuring the current from the reaction — is the textbook example; immunosensors and DNA sensors follow the same logic. DNA–protein interactions are studied by the electrophoretic mobility shift assay (EMSA/gel shift), in which a protein bound to a labelled DNA fragment migrates more slowly, by DNase I footprinting (the bound protein protects its site from cleavage) and, genome-wide, by ChIP (Chapter 2). Protein–protein interactions are studied by the yeast two-hybrid system (bait and prey fused to the two halves of a transcription factor, so an interaction switches on a reporter), co-immunoprecipitation, pull-downs and FRET/SPR. Genomics- and proteomics-based approaches — reading the whole set of genes or proteins at once — are developed in Chapter 2.
Key takeaways
- Type II restriction enzymes cut palindromic sites leaving sticky or blunt ends; T4 ligase joins them, and dephosphorylating the vector suppresses self-ligation. Sticky ends give efficient, directional cloning.
- Vectors scale by insert size: plasmids (~10 kb) with blue–white screening, phage (~9–23 kb), cosmids (~45 kb), BACs (~100–300 kb), YACs (megabases). An expression vector adds a promoter, RBS, terminator and often a tag.
- A cDNA library (from mRNA) holds only exons and is used to express eukaryotic genes in bacteria; library coverage follows N = ln(1 − P)/ln(1 − f). Gibson and Gateway allow restriction-site-independent assembly.
- PCR amplifies a primer-bounded target by 2^n per n cycles; RT-PCR starts from RNA and qPCR quantifies by Ct. Sanger uses chain-terminating ddNTPs; NGS sequences millions of fragments in parallel.
- Southern blots DNA, northern RNA (western protein); RFLP reads site/length changes via Southern, RAPD uses arbitrary primers. CRISPR–Cas9 cuts at a gRNA target by a PAM; NHEJ knocks out, HDR with a donor knocks in.
Practice questions (16)
Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.
A type II restriction endonuclease is especially useful for cloning because it
Show answer
Answer: A — cuts at a specific, defined recognition site at or very near that site
Type II enzymes cut at or immediately beside a defined, usually palindromic site and need only Mg²⁺, so the fragments they make are reproducible and defined — exactly what cloning requires. Type I and III enzymes cut away from their site (and need ATP and other cofactors), giving unpredictable fragments.In an ideal PCR a single double-stranded target molecule is amplified for 20 cycles with efficiency 1. The number of target molecules is ______ (give the value of 2 raised to 20).
Numerical answer — type the value.
Show answer
Answer: 1048576
Each cycle doubles the target, so after n cycles one molecule becomes 2^n. For n = 20 this is 2^20 = 1 048 576. (In a real PCR only the specific product bounded by both primers accumulates as 2^n after the first few cycles, and efficiency is below 1.)In pUC-based blue–white screening on X-gal, a bacterial colony carrying a plasmid with an insert in the polylinker appears
Show answer
Answer: A — white, because the insert disrupts the lacZ-alpha fragment and no functional beta-galactosidase is made
An insert in the polylinker interrupts the lacZ-alpha coding sequence, so complementation fails, no active beta-galactosidase forms, and X-gal is not cleaved to the blue product — a white colony. Empty vector keeps lacZ-alpha intact and turns blue. White colonies are the recombinants to pick.A human protein is to be produced in E. coli. The coding sequence is best obtained as
Show answer
Answer: A — cDNA made by reverse transcription of its mRNA, because it lacks introns
E. coli cannot splice, so a genomic clone with introns would not give the correct protein. cDNA, reverse-transcribed from mature mRNA, contains only the exons and so encodes the continuous coding sequence a bacterium can translate directly.In the three steps of a PCR cycle — denaturation, annealing and extension — the annealing temperature is chosen
Show answer
Answer: A — a few degrees below the melting temperature of the primers
Annealing is set just below the primers’ Tm so they bind their complementary target specifically; too low a temperature lets them bind mismatched sites and make non-specific products. 72 °C is the extension step and denaturation is around 95 °C.Sanger dideoxy sequencing works because a dideoxynucleotide, once incorporated,
Show answer
Answer: A — cannot form the next phosphodiester bond, because it lacks a 3-prime hydroxyl
A ddNTP base-pairs normally but has no 3-prime hydroxyl, so no further nucleotide can be added and synthesis terminates at that base. Running fragments of every termination length and reading the terminating base gives the sequence.Northern blotting is used to
Show answer
Answer: A — detect a specific RNA and measure a transcript’s size and abundance
Northern blotting separates RNA, transfers it and probes it, so it measures whether and how much a gene’s transcript is present and its size — a readout of expression. DNA is Southern, protein is western, and amplification is PCR.After a CRISPR–Cas9 double-strand break is repaired by non-homologous end joining (NHEJ) with no donor template, the usual result is
Show answer
Answer: A — small insertions or deletions that frameshift and knock the gene out
NHEJ rejoins the broken ends imprecisely, leaving small indels that usually shift the reading frame and abolish the protein — a knock-out. A precise change (knock-in) needs a donor template and homology-directed repair, which NHEJ is not.The yeast two-hybrid system is used to detect
Show answer
Answer: A — protein–protein interactions
In the two-hybrid system a bait protein and a prey protein are fused to the DNA-binding and activation domains of a transcription factor; only if bait and prey interact are the domains brought together to switch on a reporter gene. It reports protein–protein interaction, not DNA sites, RNA size or primer Tm.A genomic library is made with an average insert of 20 kb from a genome of 4000 kb. The number of independent clones needed for a 99% probability that any sequence is represented is ______ (round off to the nearest whole number; use N = ln(1 − P)/ln(1 − f)).
Numerical answer — type the value.
Show answer
Answer: 919
f = 20/4000 = 0.005 and P = 0.99, so N = ln(1 − 0.99)/ln(1 − 0.005) = ln(0.01)/ln(0.995) = (−4.6052)/(−0.0050125) = 918.7, about 919. Using 0.995 or forgetting the logs gives far too few clones.Which of the following vector–capacity pairings are correct? Select all that apply.
Show answer
Answer: A — Plasmid — up to roughly 10 kb; B — Cosmid — up to roughly 45 kb; C — BAC — of the order of 100–300 kb
Ordinary plasmids take small inserts (~10 kb), cosmids about 45 kb and BACs hundreds of kb; megabase inserts need YACs, not plasmids, so the pairing of plasmids with megabase capacity is wrong. The larger the capacity, the fewer clones a genome needs.Before ligating an insert, a linearised plasmid vector is treated with alkaline phosphatase. The purpose is to
Show answer
Answer: A — remove the vector’s 5-prime phosphates so it cannot re-ligate to itself without an insert
Ligase needs a 5-prime phosphate to form the phosphodiester bond. Removing the vector’s 5-prime phosphates means the two vector ends cannot join each other, so the vector can only close by using the insert’s phosphates — which suppresses empty self-ligated background. It does not blunt, methylate or add phosphates.In real-time quantitative PCR, a sample with more starting template gives a threshold cycle (Ct) that is
Show answer
Answer: A — lower, because the fluorescence crosses the threshold after fewer cycles
More starting template reaches the detection threshold sooner, so its Ct is lower; Ct falls by about one cycle for each doubling of input. A higher Ct means less starting material. This inverse relationship is what lets qPCR quantify a sample.Which statements comparing RFLP and RAPD for DNA fingerprinting are correct? Select all that apply.
Show answer
Answer: A — RFLP is highly reproducible and codominant but laborious and needs good-quality DNA; B — RAPD uses single short primers of arbitrary sequence and needs no prior sequence information; C — RAPD is a dominant marker and more sensitive to reaction conditions
RFLP is reproducible, codominant and Southern-blot-based (no arbitrary-primer PCR); RAPD uses single arbitrary primers, needs no sequence information, and is dominant and condition-sensitive. The statement that RFLP requires PCR with arbitrary primers wrongly assigns RAPD’s arbitrary-primer PCR to RFLP.In a modern whole-plasmid site-directed mutagenesis kit, after amplifying the plasmid with mutagenic primers the reaction is treated with DpnI. DpnI is used because it
Show answer
Answer: A — cuts only methylated DNA, so it destroys the parental template but not the newly synthesised mutant
The parental plasmid was grown in E. coli and so is Dam-methylated, whereas the plasmid synthesised in vitro with the mutagenic primers is not. DpnI cuts only methylated (and hemimethylated) DNA, so it selectively destroys the template and leaves the unmethylated mutant to be recovered.Gibson assembly and Gateway cloning are valued mainly because they
Show answer
Answer: A — can join or move DNA without depending on unique restriction sites
Both are recombination- or overlap-based methods that assemble constructs without requiring convenient unique restriction sites — Gibson joins overlapping fragments in one isothermal reaction, Gateway uses phage att-site recombination. This is what makes seamless, scar-free, multi-fragment construction routine.