Genetics and Molecular Biology · Unit 5: Tools and genomes

Recombinant DNA, Cloning and PCR

How do you get millions of copies of one gene, and how do you know you have the right one? This lesson covers the two routes: cutting DNA with restriction enzymes, joining it into a plasmid with ligase and growing it in E. coli, or copying one region directly by PCR. It then shows how qPCR turns the cycle at which a signal appears into a measure of starting amount, and how gels and Southern, Northern and Western blots read out the result. Afterwards you can plan a cloning experiment, calculate a qPCR fold change and choose the controls that make each result believable.

Watch first Copies of Copies, Doubling Each Cycle · 2:37

In 1985 a team at Cetus Corporation reported raising the number of copies of one short stretch of the human beta-globin gene about 220,000-fold, in a tube, with no living cell involved.

Before then, the usual way to get many copies of one piece of DNA was to cut it out, paste it into a small circle of DNA that bacteria copy, and let the bacteria grow. Both routes are still in daily use. This lesson takes them in order, works through the arithmetic that makes PCR powerful and quantitative, and ends with how you read the result and which controls make it believable.

Cut and paste: restriction enzymes and ligase

Bacteria are attacked by phages, viruses that inject their DNA. Many bacteria defend themselves with a restriction enzyme, which cuts DNA at one short sequence, paired with a methylase that marks the same sequence in the cell’s own DNA so that it is spared. Unmarked phage DNA gets cut.

The restriction enzymes used in the laboratory are type II enzymes. Each recognizes a defined sequence, in many cases a palindrome of four to eight base pairs, and cuts at a fixed position in or near it. Palindrome here means the site reads the same 5’ to 3’ on both strands. EcoRI, from E. coli, recognizes 5’-GAATTC-3’ and cuts each strand between the G and the A. The two cuts are four bases apart, so each end carries a single-stranded 5’ overhang reading AATT. Any EcoRI end can pair with any other EcoRI end through those four bases, which is why such ends are called sticky, or cohesive. Other enzymes, such as SmaI at CCCGGG, cut both strands at the same point and leave blunt ends.

Think of a jigsaw cutter that always cuts the same tab shape: any two pieces cut by it fit, but only loosely. DNA ligase makes the join permanent by forming the missing phosphodiester bond between a 3’-OH and a 5’-phosphate on each strand. T4 DNA ligase can also join blunt ends in vitro.

The first recombinant molecules were made entirely in the test tube. In 1972 David Jackson, Robert Symons and Paul Berg joined lambda phage genes and the E. coli galactose operon to the DNA of the virus SV40. The same year Janet Mertz and Ronald Davis showed that EcoRI ends are complementary and that restriction enzyme plus ligase can make hybrid molecules. The next step was to have such a molecule copied by a living cell.

Into bacteria: finding the colony that carries your insert

A plasmid is a small circle of DNA that a bacterium copies separately from its chromosome (see bacterial and phage genetics). A cloning vector is a plasmid reduced to three working parts. An origin of replication is the site where the host’s replication machinery starts copying; without it the plasmid is not copied and is lost as cells divide. A selectable marker, such as a gene for ampicillin resistance, lets only plasmid-bearing cells grow. A cloning site holds unique restriction sites where the insert goes.

In 1973 Stanley Cohen, Annie Chang, Herbert Boyer and Robert Helling joined restriction fragments from separate plasmids, put the products into E. coli and recovered plasmids that replicated and carried properties of both parents. Berg’s group had made recombinant DNA first; the Cohen and Boyer collaboration first showed a recombinant plasmid working as a replicon in a cell.

Getting the plasmid in is transformation: uptake of naked DNA by a bacterial cell. Cells are made competent with cold calcium chloride or by electroporation, and only a fraction take up a plasmid. Bacterial transformation is a different process from transfection, which is putting DNA into animal cells, and from oncogenic transformation, which is a cell becoming cancer-like.

Now the two questions every cloning experiment must keep apart. Selection asks whether a cell has a plasmid: on ampicillin, cells without one die. But a vector that closed up without an insert also confers resistance. Screening asks which plasmid it is. In blue-white screening the cloning site sits inside lacZ-alpha, a short piece of the beta-galactosidase gene from the lac operon. The host strain makes the rest of the enzyme, and the two pieces together are active. Active enzyme turns the colorless substrate X-gal blue. An insert in the cloning site breaks lacZ-alpha, so recombinant colonies stay white. White is a strong hint, not proof: some white colonies carry no insert. The final checks are a restriction digest or colony PCR, then sequencing across both junctions. Even then a correct plasmid need not express its insert; that needs a promoter and the other parts of an expression construct (see hacking the cell).

PCR: copying one region a billion times

PCR needs a template, two primers, the four dNTPs (see nucleotide metabolism) and a DNA polymerase. Polymerase can only add nucleotides to the 3’-OH of a primer paired to a template, as you saw in DNA replication. Each cycle has three steps:

  1. Denature at about 94 to 98 °C, so the two strands separate.
  2. Anneal at a lower temperature, set a few degrees below the primers’ melting temperature, so each primer pairs with its complementary site.
  3. Extend, typically at 72 °C for Taq polymerase, so each primer is lengthened 5’ to 3’.

Primer design follows from the polymerase rule. The two primers bind opposite strands, with their 3’ ends pointing toward each other, so each new strand contains a binding site for the other primer. Then products of one cycle become templates for the next, like a photocopier whose copies can be copied. Primers are commonly 15 to 30 nucleotides, the pair should have matched melting temperatures, and each should avoid pairing with itself or folding into a hairpin. Tools such as Primer-BLAST check a pair against a sequence database for unintended targets.

Exact-length products do not appear at once. In cycle 1, copies of a genomic template start at a primer but run past the other site. Molecules bounded by both primers first appear in cycle 3, and after n cycles there are 2^n - 2n of them per starting duplex. By cycle 30 they are nearly all of the product.

PCR cycle plotter: product against cycle number⚙ original · interactive
Copies after the chosen cycles, ideal N0 x (1 + E)^n
Copies in the plotted reaction (plateau level is a sketch)
Ct expected from the formula
Ct observed: first whole cycle at the threshold

Keep the rule: a lower Ct means more starting template, and at 100% efficiency one cycle earlier means twice as much; a plateau hides the starting amount at the end but not at the threshold. The plateau level and threshold height are relative units, a sketch of the logic.

Worked problem (a). You start with 10 double-stranded template molecules. Assuming perfect doubling every cycle, how many copies are there after 30 cycles?

N = N0 x (1 + E)^n, with E = 1 for perfect doubling

  1. Each cycle multiplies by 2, so 30 cycles multiply by 2^30 = 1,073,741,824.
  2. N = 10 x 1,073,741,824 = 10,737,418,240, or about 1.07 x 10^10.

A real reaction stops short. Amplification is exponential only in the early cycles; then it slows and reaches a plateau. In one model system, reactions with different starting amounts converged on the same final level, so a bright endpoint band does not tell you how much template you began with.

The first published PCR, in 1985, used the Klenow fragment of E. coli DNA polymerase I, which heat destroys, so fresh enzyme went in every cycle. In 1988 the Cetus group reported PCR with the polymerase of Thermus aquaticus, which survives denaturation. Taq lacks a 3’ to 5’ proofreading exonuclease, so its copies carry more errors; in one 1991 reporter assay, products made with Pfu polymerase carried fewer than a tenth as many mutations. Kary Mullis shared the 1993 Nobel Prize in Chemistry for conceiving PCR.

Counting with qPCR

Two names are easily confused. RT-PCR is reverse-transcription PCR: reverse transcriptase first copies RNA into complementary DNA (cDNA), which is then amplified. Real-time quantitative PCR (qPCR) measures product during cycling with a fluorescent dye that binds double-stranded DNA or a sequence-specific probe. RT-qPCR combines the two to measure mRNA.

The readout is the threshold cycle, Ct: the cycle at which fluorescence crosses a set threshold during the exponential phase. A lower Ct means more starting template, because fewer doublings are needed to reach the same amount of product. At 100% efficiency, one cycle earlier means twice as much template.

Worked problem (b). Both reactions are about 100% efficient. Target gene Ct is 24.0 in treated cells and 26.5 in control; reference gene Ct is 18.0 in treated and 18.5 in control.

fold change = 2^(-ΔΔCt), where ΔCt = Ct(target) - Ct(reference)

  1. ΔCt treated = 24.0 - 18.0 = 6.0.
  2. ΔCt control = 26.5 - 18.5 = 8.0.
  3. ΔΔCt = 6.0 - 8.0 = -2.0.
  4. Fold change = 2^2 = 4. Target mRNA is about 4-fold higher in treated cells.

The method assumes equal efficiencies near 100% for both assays and a reference gene shown not to change in this experiment.

Reading the result: gels and blots

DNA’s phosphate backbone is negatively charged, so in an agarose gel it moves toward the positive electrode. The gel is a mesh, and smaller linear fragments slip through faster. A size ladder of known fragments in a neighboring lane turns position into length. Uncut circular plasmid runs at a different position from the same plasmid cut to a line. A band of the right size is evidence about size, not identity.

Blots add identity. In Southern’s 1975 method, fragments in the gel are denatured and transferred to a membrane, which is then soaked in a labeled probe; only fragments that pair with it light up. The Northern blot applies the same idea to RNA. The Western blot transfers proteins separated by SDS-PAGE and detects one with an antibody (covered in protein purification techniques).

Each readout needs its controls:

  • a size ladder on every gel;
  • a no-template control for every PCR, to detect contamination, and a no-RT control for RNA, to detect genomic DNA;
  • a positive control, so a negative result means absent rather than failed;
  • a loading control on blots, such as a reprobe for a stable RNA or a total-protein stain, read within its linear range.

Design problem

You cut a 5.0 kb vector and a 1.2 kb insert with EcoRI and BamHI, ligate, transform and select on ampicillin. Insert plus vector gives 200 colonies; vector alone with ligase gives 180. Interpret the result and redesign the experiment.

One way to do it

Interpret. An EcoRI end (AATT overhang) cannot pair with a BamHI end (GATC overhang), so a vector cut by both enzymes cannot close on itself. Colonies from vector alone must come from vector that was cut by only one enzyme, whose matching ends re-ligate, or not cut at all, since uncut plasmid transforms well. That background is 180 of 200, so most colonies on the insert plate probably lack the insert.

Redesign.

  1. Digest the vector longer or with more enzyme, and check each single digest.
  2. Gel-purify the cut vector away from uncut circles. Singly and doubly cut vector run at almost the same size, so this step alone is not enough.
  3. Dephosphorylate the vector with alkaline phosphatase. Without 5’-phosphates it cannot re-ligate, while the insert still supplies phosphates for each junction.
  4. Add controls: cut vector without ligase, to measure uncut and singly cut carry-over; and uncut plasmid, to measure transformation efficiency.
  5. Screen colonies by colony PCR or by EcoRI plus BamHI digest, expecting bands of 5.0 and 1.2 kb, then sequence across both junctions.

How we measure it

Restriction-ligation cloning

Vector and insert are cut with restriction enzymes to give compatible ends, joined with DNA ligase, and introduced into E. coli by transformation. Antibiotic selection keeps cells that carry a plasmid; screening, digestion and sequencing show which plasmid it is.

PCR and real-time quantitative PCR

Two primers and a heat-stable polymerase copy the region between them in repeated cycles of denaturing, annealing and extension. Watching fluorescence each cycle gives a Ct value that falls as the starting amount of template rises.

Southern blotting

Restriction fragments separated on an agarose gel are denatured and transferred to a membrane, then hybridized with a labeled probe. Bands show which fragment sizes carry the probe's sequence.

Agarose gel electrophoresis

DNA is loaded into a gel and pulled toward the positive electrode by its negatively charged backbone. Smaller linear fragments move faster, and a size ladder in a neighboring lane turns position into an estimate of length.

Finish line

Test yourself

8 questions

Practice and MCAT-style, with explained answers.

Next · Lesson 23

Sequencing and Genomes

Two drafts of the human genome appeared in 2001. Why did a gapless one take until 2022?

Continue →

← Lesson 21 · Cancer as a Genetic Disease

EcoRI bound to its DNA site — the enzyme that makes sticky ends
EcoRI bound to its DNA site — the enzyme that makes sticky ends ● LIVE · RCSB PDB 1ERI
open on RCSB PDB ↗
Loading live structure 1ERI from RCSB PDB…

Two identical subunits clasp the double helix, one for each half of the palindrome. Each subunit cuts one strand, and the two cuts are four base pairs apart, which is what leaves the AATT overhangs.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Taq DNA polymerase — the heat-stable enzyme of PCR
Taq DNA polymerase — the heat-stable enzyme of PCR ● LIVE · RCSB PDB 1TAQ
open on RCSB PDB ↗
Loading live structure 1TAQ from RCSB PDB…

Find the polymerase domain shaped like a cupped right hand, and the separate 5' to 3' exonuclease domain at the other end. The domain that would proofread in E. coli polymerase I is present in shape but lacks the 3' to 5' exonuclease activity.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Taq polymerase large fragment with primer, template and incoming nucleotide — one step of extension
Taq polymerase large fragment with primer, template and incoming nucleotide — one step of extension ● LIVE · RCSB PDB 3KTQ
open on RCSB PDB ↗
Loading live structure 3KTQ from RCSB PDB…

The primer's 3' end sits in the active site with the template strand running past it. The incoming nucleotide pairs with the next template base, the step repeated thousands of times in every extension.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Hamilton O. Smith

Hamilton O. Smith

With Kent Wilcox, purified and characterized a restriction enzyme from Haemophilus influenzae Rd, an enzyme that cuts DNA at a defined sequence, which made sequence-specific cutting a laboratory tool.

By the late 1960s it was known that some bacteria destroy foreign DNA that enters them while sparing their own, a pattern called restriction. What was missing was an enzyme that did the cutting at a fixed address. In 1970 Hamilton Smith and Kent Wilcox reported in the Journal of Molecular Biology the purification and characterization of a restriction enzyme from Haemophilus influenzae strain Rd. A companion paper from Smith's laboratory with Thomas Kelly worked out the sequence at which it cuts. An enzyme that always cuts at the same short sequence turns a long DNA molecule into a reproducible set of pieces. Daniel Nathans, also at Johns Hopkins, soon used such an enzyme to cut the DNA of the virus SV40 into fragments that could be ordered into a map. Smith shared the 1978 Nobel Prize in Physiology or Medicine with Werner Arber and Nathans. Every restriction digest in this lesson, from cloning to the Southern blot, uses the kind of enzyme his work made available.

✓ verbatim Smith & Wilcox, J. Mol. Biol. 51:379 (1970); Nobel Prize in Physiology or Medicine 1978, nobelprize.org.
Stanley N. Cohen and Herbert W. Boyer

Stanley N. Cohen and Herbert W. Boyer

With Annie Chang and Robert Helling, joined restriction fragments of separate plasmids in vitro and showed that the recombinant plasmids replicated in E. coli and carried properties and sequences of both parents.

Cohen's laboratory at Stanford studied how bacterial plasmids acquire antibiotic-resistance genes. In 1972, with Annie Chang and Leslie Hsu, it reported that E. coli treated with calcium chloride could take up purified plasmid DNA and become resistant to antibiotics. Boyer's laboratory at the University of California, San Francisco, worked on the restriction enzyme EcoRI, whose ends are cohesive. Together, in a PNAS paper published in November 1973 with Chang and Robert Helling, they cut two different plasmids, joined fragments in vitro, transformed E. coli and recovered plasmids that replicated and carried genetic properties and sequences from both parents. Paul Berg's group had already joined DNA from different sources in a test tube in 1972; the 1973 work showed that a recombinant plasmid could be propagated in living bacteria. In 1974 a paper with Cohen and Boyer among its authors reported frog ribosomal DNA carried in an E. coli plasmid. That made it possible to grow any piece of DNA you could cut out.

✓ verbatim Cohen, Chang, Boyer & Helling, Proc. Natl. Acad. Sci. USA 70:3240 (1973); Cohen, Proc. Natl. Acad. Sci. USA 110:15521 (2013).
Kary B. Mullis

Kary B. Mullis

Conceived the polymerase chain reaction at Cetus Corporation, in which two primers and repeated rounds of copying amplify the DNA between them; shared the 1993 Nobel Prize in Chemistry with Michael Smith.

Mullis worked at Cetus Corporation, where the laboratory next door, Henry Erlich's, was developing ways to detect point mutations. In his Nobel lecture he describes how his own attempt to read a single base in human DNA, hard because one gene is a tiny fraction of the genome, grew into PCR: use two primers that face each other, so that the products of one round of copying become templates for the next. The first published application came in 1985 from Randall Saiki and colleagues at Cetus, Mullis among them. They amplified a stretch of the human beta-globin gene about 220,000-fold and used it to tell the normal and sickle-cell alleles apart. That reaction used the Klenow fragment of E. coli DNA polymerase I, which heat destroys, so fresh enzyme was added every cycle. In 1988 the same group reported PCR with a heat-stable polymerase from Thermus aquaticus, which needed no replenishment. Mullis received the 1993 Nobel Prize in Chemistry. PCR made it possible to study a gene from very small samples, down to a single human sperm.

✓ verbatim Mullis, Nobel Lecture, 8 December 1993; Saiki et al., Science 230:1350 (1985); Saiki et al., Science 239:487 (1988).
Edwin M. Southern

Edwin M. Southern

Developed the transfer of gel-separated DNA fragments to a filter for hybridization with a labeled probe, so that a specific sequence can be found among thousands of fragments.

A restriction digest of genomic DNA run on a gel gave a smear, because there were too many fragments to see one. Southern, at the MRC Mammalian Genome Unit in Edinburgh, wanted to identify the fragments complementary to a given RNA without losing the separation the gel had achieved. His 1975 paper in the Journal of Molecular Biology describes the method. The DNA in the gel is denatured, then carried by capillary flow onto a cellulose nitrate filter laid against it, so the filter holds a copy of the gel's pattern. The filter is soaked in radioactive RNA, washed and exposed to film, and only the fragments that pair with the probe show up. He demonstrated it with fragments complementary to ribosomal RNA in DNA from E. coli, the frog Xenopus laevis and several mammals. The method gave both the size of a fragment and its identity, and it became the model for the RNA and protein blots that followed.

✓ verbatim Southern, J. Mol. Biol. 98:503 (1975); Southern, Nature Protocols 1:518 (2006).
Test yourself
Question 1 of 8 Score 0 / 0

EcoRI recognizes 5'-GAATTC-3' and cuts each strand between the G and the A. What ends does it leave?
Sources & Integrity

Every structure, slide, simulation, and clip on this page is listed below with the source it came from. None of the 3D structures is drawn by hand or by AI: each one loads by accession ID from RCSB PDB or PubChem. This panel is generated from the lesson's own data file, so what it lists is what the page loads.

Molecular structures live data

  • EcoRI bound to its DNA site — the enzyme that makes sticky ends — RCSB PDB 1ERI · coordinates CC0 · EcoRI restriction endonuclease in complex with a DNA duplex containing GAATTC, X-ray crystal structure
  • Taq DNA polymerase — the heat-stable enzyme of PCR — RCSB PDB 1TAQ · coordinates CC0 · Full-length DNA polymerase I from Thermus aquaticus, X-ray crystal structure (Kim et al., Nature, 1995)
  • Taq polymerase large fragment with primer, template and incoming nucleotide — one step of extension — RCSB PDB 3KTQ · coordinates CC0 · Klentaq1 ternary complex with primer-template DNA and ddCTP, X-ray crystal structure (Li, Korolev & Waksman, EMBO J., 1998)

Simulations

  • PCR cycle plotter: product against cycle number · Original interactive — Biochemistrypedia (Deppmann Lab).

Scientist profiles & quotes

  • Hamilton O. Smith — Smith & Wilcox, J. Mol. Biol. 51:379 (1970); Nobel Prize in Physiology or Medicine 1978, nobelprize.org.
  • Stanley N. Cohen and Herbert W. Boyer — Cohen, Chang, Boyer & Helling, Proc. Natl. Acad. Sci. USA 70:3240 (1973); Cohen, Proc. Natl. Acad. Sci. USA 110:15521 (2013).
  • Kary B. Mullis — Mullis, Nobel Lecture, 8 December 1993; Saiki et al., Science 230:1350 (1985); Saiki et al., Science 239:487 (1988).
  • Edwin M. Southern — Southern, J. Mol. Biol. 98:503 (1975); Southern, Nature Protocols 1:518 (2006).

Quotes are verbatim from the cited source. Profile text is original to this site and was checked against the sources cited.

Animated explainer

Copies of Copies, Doubling Each Cycle · narration: an AI narrator voice (Gemini, Charon) reading narration written from this lesson's text · animation: AI-generated in code (Manim) by Claude, a schematic with no drawn molecular structures, plots computed from their equations · captions from the narration script.

Images & figures

This lesson carries no raster figures. Every structure is rendered live from RCSB PDB or PubChem, and every plot is drawn from its equation in the browser.

integrity Machines check the mechanics. The build confirms that every local file exists, that every question's answer key points at a real choice, and that any AI-generated scientific figure names the person who checked it; a weekly job re-resolves every accession ID against its database. Questions are original; MCAT-style items are written for this site and are not real AAMC questions. Banned sources (the Tymoczko textbook, publisher decks, copyrighted figures) are not used.

Like I'm five

Some bacteria have scissors that cut DNA only at one special spelling. Scientists use those scissors to cut out a gene and glue it into a tiny ring that bacteria copy whenever they divide. There is also a faster trick, PCR, that copies one piece of DNA again and again in a tube, doubling it each time, until there is enough to see.
High school

Restriction enzymes come from bacteria, where they cut invading virus DNA at short, specific sequences. Many leave short single-stranded tails that stick to any other end cut by the same enzyme, and DNA ligase seals the join. Joining a gene into a plasmid and putting it into E. coli lets the bacteria make copies as they grow; an antibiotic-resistance gene on the plasmid lets only cells with a plasmid survive. PCR copies one region in a test tube: heat separates the strands, two short primers stick on either side of the region, and a heat-stable polymerase copies it. Each cycle can double the amount, so 30 cycles can make about a billion copies from one, until the reaction runs out of steam.
College

Type II restriction enzymes cut at defined, often palindromic, sites of four to eight base pairs; EcoRI cuts GAATTC to leave 5' AATT overhangs, and others leave blunt ends. Ligase forms phosphodiester bonds between a 3'-OH and a 5'-phosphate. A vector needs an origin, a selectable marker such as ampicillin resistance and a cloning site; selection keeps transformed cells, while screening, for example by insertional inactivation of lacZ-alpha on X-gal, suggests which carry an insert, and digestion or sequencing confirms it. PCR repeats denaturation, annealing of two inward-facing primers on opposite strands, and extension. At 100% efficiency the target doubles each cycle, N = N0 x 2^n, until the reaction plateaus. In real-time qPCR the threshold cycle Ct falls by about one for each doubling of starting template, and with equal, near-100% efficiencies the fold change is 2 to the minus delta delta Ct. RT-PCR adds a reverse-transcription step to start from RNA. Gels sort by size; Southern, Northern and Western blots add a probe or antibody to identify DNA, RNA or protein.
Graduate

The constant-efficiency model, N_c = N_0(1 + E)^c, holds only in the early cycles; product accumulation then slows, and experiments have implicated polymerase binding to accumulated product in one system and assay-dependent primer depletion in others. Because inputs converge at the plateau, quantification uses the cycle at which fluorescence crosses a threshold, and either a standard curve of known inputs (absolute) or the 2^-ΔΔCt method (relative), which assumes equal efficiencies near 100% and a reference gene validated for the experiment. Vandesompele and colleagues found large errors from single reference genes in a substantial fraction of human tissue samples, which is why geometric means of several validated references are used. MIQE guidelines (2009) and their précis (2010) ask for no-RT and no-template controls, and the 2025 revision asks for prediction intervals, detection limits and dynamic ranges for each target. On the cloning side, selection and color are phenotypes, not sequence: some white colonies carry no insert, so junction sequencing remains the decisive check. An open question for any qPCR study is how much of the variance between replicates comes from efficiency differences that a single assumed E hides.
Fresh enzyme every cycle

The first published PCR used the Klenow fragment of E. coli DNA polymerase I. Each heating step to separate the strands destroyed it, so fresh enzyme was added after every denaturation, cycle after cycle. The fix came from a bacterium that Thomas Brock and Hudson Freeze had described in 1969, Thermus aquaticus, which grows in hot water. Its DNA polymerase survives the denaturing step. In 1988 Saiki and colleagues reported that with this enzyme no polymerase had to be added during cycling, and that segments up to 2,000 base pairs were readily amplified in their study.

— Saiki et al., Science 230:1350 (1985); Saiki et al., Science 239:487 (1988); Brock & Freeze, J. Bacteriol. 98:289 (1969).
A compass of blots

Only one blot is named after a person. Edwin Southern published the DNA transfer in 1975. In 1977 James Alwine, David Kemp and George Stark transferred gel-separated RNA to paper and detected it with probes, and that RNA method came to be called the Northern blot. Protein transfers to membranes were published by several groups in 1979, and W. Neal Burnette used the name western blotting in the title of his 1981 paper. The names Northern and Western began as jokes on Southern's name.

— Southern, J. Mol. Biol. 98:503 (1975); Alwine, Kemp & Stark, Proc. Natl. Acad. Sci. USA 74:5350 (1977); Burnette, Anal. Biochem. 112:195 (1981).
Who made recombinant DNA first?

In 1972 David Jackson, Robert Symons and Paul Berg reported joining lambda phage genes and the E. coli galactose operon to SV40 DNA in a test tube, using complementary single-nucleotide tails added by terminal transferase. Peter Lobban, a graduate student in Dale Kaiser's Stanford laboratory, independently conceived enzymatic DNA joining and published it with Kaiser in 1973. Cohen and Boyer's 1973 work was the first to propagate such a plasmid in bacteria. Accounts of how much credit belongs where still differ.

— Jackson, Symons & Berg, Proc. Natl. Acad. Sci. USA 69:2904 (1972); Berg & Mertz, Genetics 184:9 (2010); Cohen, Proc. Natl. Acad. Sci. USA 110:15521 (2013).

 ↑ Top