The meristem — both apical and axillary.

Why. Even if a plant is infected with a virus, the meristem is free of virus. Hence one can remove the meristem and grow it in vitro to obtain virus-free plants.

Why the meristem escapes infection. It is the actively dividing growing region. Viruses spread mainly through the vascular tissue, which is not yet differentiated in the meristem, and the rapid rate of cell division there tends to outrun viral replication, so the newest cells are formed ahead of the infection.

What makes the technique work. A meristem tip is a tiny fragment, not a plant. It can be grown into a complete plant only because of totipotency — the capacity to generate a whole plant from any cell or explant — and only under sterile conditions in a special nutrient medium.

Where it has been done. Scientists have succeeded in culturing meristems of banana, sugarcane and potato. All three are propagated vegetatively, so without this technique a virus would be carried indefinitely from one generation of planting material to the next.

The major advantage is that it is possible to achieve propagation of a very large number of plants in very short durations — thousands of plants from a single parent, by tissue culture.

Three further advantages that follow from it.

(i) Genetic uniformity. Each of these plants is genetically identical to the original plant from which it was grown, that is they are somaclones. The whole crop therefore ripens together, reaches the same size and responds the same way to treatment, which is what makes commercial-scale growing and marketing practicable. Many important food plants like tomato, banana and apple have been produced on a commercial scale using this method.

(ii) Elite parents can be multiplied. A single outstanding plant — or a single successful transgenic plant — can be turned into a plantation, which no seed-based method could do since sexual reproduction reshuffles the genotype.

(iii) Disease-free stock. The same technique can recover healthy plants from diseased ones by culturing the virus-free meristem.

The cost of that advantage. Because every plant is genetically identical, every plant has identical weaknesses, and a new pathogen able to infect one can infect the entire plantation.

The nutrient medium must supply everything the explant would normally obtain from the rest of the plant, because the fragment has no roots and no photosynthetic supply of its own. Its components are:

One condition matters as much as the composition: the explant must be grown under sterile conditions. A medium rich enough to feed a plant cell will feed a contaminating bacterium or fungus far faster, and the culture would be lost.

The correct answer is (c) — the toxin is inactive.

Explanation. The Bt toxin protein exists as inactive protoxins. Once an insect ingests the inactive toxin, it is converted into the active form of toxin, because the alkaline pH of the insect gut solubilises the crystals. The activated toxin then binds to the surface of the midgut epithelial cells and creates pores that cause cell swelling and lysis, and eventually the death of the insect. Inside the bacterium there is no alkaline gut, so the protoxin is never activated and does no harm.

Why the others are wrong. (a) The bacterium is not resistant; resistance would mean the active toxin is present but ineffective. (b) The protein is not immature or half-built — it is a complete protoxin awaiting chemical activation, exactly as a pro-enzyme is a complete molecule awaiting processing. (d) There is no special sac; the toxin is held as protein crystals formed during a particular phase of the bacterium's growth.

The same fact explains a practical point. Mammals have a strongly acidic stomach, not an alkaline gut, so the protoxin is digested as an ordinary protein rather than activated.

Definition. Transgenic bacteria are bacteria whose DNA has been manipulated to possess and express an extra, foreign gene, introduced by recombinant DNA technology.

The example: E. coli producing human insulin.

Insulin consists of two short polypeptide chains, chain A and chain B, linked together by disulphide bridges. In mammals it is synthesised as a pro-hormone containing an extra stretch called the C peptide, which is not present in mature insulin and is removed during maturation. The main challenge for producing insulin by rDNA techniques was therefore getting insulin assembled into a mature form, since a bacterium cannot carry out that processing.

In 1983, Eli Lilly, an American company, prepared two DNA sequences corresponding to the A and B chains of human insulin and introduced them into plasmids of E. coli to produce the insulin chains. Chains A and B were produced separately, extracted, and combined by creating disulfide bonds to form human insulin.

Why it mattered. Insulin used for diabetes was earlier extracted from the pancreas of slaughtered cattle and pigs, and insulin from an animal source caused some patients to develop allergy or other types of reactions to the foreign protein. Human insulin is made in bacteria yet its structure is absolutely identical to that of the natural molecule, so it does not induce unwanted immunological responses — and a large quantity of bacteria can be grown to make as much insulin as is needed.

Advantages. Genetic modification has:

• made crops more tolerant to abiotic stresses — cold, drought, salt and heat;
• reduced reliance on chemical pesticides, through pest-resistant crops such as Bt cotton, in effect creating a bio-pesticide;
• helped to reduce post-harvest losses;
• increased the efficiency of mineral usage by plants, which prevents early exhaustion of the fertility of the soil;
• enhanced the nutritional value of food, as in golden rice, that is vitamin A enriched rice;
• created tailor-made plants to supply alternative resources to industries, in the form of starches, fuels and pharmaceuticals;
• increased crop yields where conventional breeding had reached its limit, since further increases in yield with existing varieties are not possible by conventional breeding.

There is also an economic advantage of real importance: for farmers in the developing world, agrochemicals are often too expensive, so a plant that defends itself costs the farmer nothing per application.

Disadvantages and concerns.

• Unpredictable ecological effects. Genetic modification of organisms can have unpredictable results when such organisms are introduced into the ecosystem. Unlike a chemical, a GM organism reproduces, spreads and can pass its genes on.
• Effects on non-target organisms. A toxin expressed in every tissue of a widely grown crop is encountered by insects that were never the target, since most Bt toxins are specific only to an insect group.
• Evolution of resistance. Continuous exposure selects for pests that survive the toxin, so the protection is not permanent.
• Loss of genetic diversity. Wide adoption of a few engineered varieties displaces the many local varieties — India alone has an estimated 200,000 varieties of rice — and that diversity is the raw material of all future breeding.
• Dependence and cost. Seed must be bought each season, which shifts the balance of power from the farmer to the seed company.
• Patents and biopiracy. The modification or usage of living organisms for public services has created problems with patents, and there is growing public anger that companies are granted patents for products that make use of biological resources long developed and used by farmers and indigenous people.
• Ethical objections. The manipulation of living organisms cannot go on without regulation, and ethical standards are required to evaluate the morality of activities that might help or harm living organisms.

The regulatory response. This is why the Indian Government has set up organisations such as GEAC, the Genetic Engineering Approval Committee, which makes decisions regarding the validity of GM research and the safety of introducing GM organisms for public services. The balance of advantage and disadvantage is not settled once for all crops; it has to be assessed case by case before release.

What they are. Cry proteins are toxic insecticidal proteins coded by the genes named cry. They are held in protein crystals that the bacterium forms during a particular phase of its growth, and they exist as inactive protoxins until an insect ingests them, when the alkaline pH of the gut solubilises the crystals and converts the protoxin to active toxin. The activated toxin binds the surface of the midgut epithelial cells and creates pores that cause cell swelling and lysis, and eventually the death of the insect.

The organism. Bacillus thuringiensis — Bt for short. Some of its strains produce proteins that kill certain insects such as lepidopterans (tobacco budworm, armyworm), coleopterans (beetles) and dipterans (flies, mosquitoes).

How man has exploited it. The Bt toxin gene has been cloned from the bacteria and expressed in plants, to provide resistance to insects without the need for insecticides; in effect this created a bio-pesticide. Specific Bt toxin genes were isolated from Bacillus thuringiensis and incorporated into several crop plants such as cotton. Examples are Bt cotton, Bt corn, rice, tomato, potato and soyabean.

Choosing the gene. The choice depends upon the crop and the targeted pest, as most Bt toxins are insect-group specific. The proteins encoded by cryIAc and cryIIAb control the cotton bollworms, while that of cryIAb controls corn borer.

Definition. Gene therapy is a collection of methods that allows correction of a gene defect that has been diagnosed in a child or embryo. Genes are inserted into a person's cells and tissues to treat a disease, and correction of a genetic defect involves delivery of a normal gene into the individual or embryo, to take over the function of and compensate for the non-functional gene. Viruses that attack their hosts and introduce their genetic material into the host cell as part of their replication cycle are used as vectors to transfer the healthy genes.

The disorder. Adenosine deaminase deficiency is caused due to the deletion of the gene for adenosine deaminase, an enzyme crucial for the immune system to function.

What existed before gene therapy. In some children ADA deficiency can be cured by bone marrow transplantation; in others it can be treated by enzyme replacement therapy, in which functional ADA is given to the patient by injection. The problem with both approaches is that they are not completely curative.

The first clinical gene therapy, 1990. It was given to a four-year-old girl with ADA deficiency. As a first step, lymphocytes from the blood of the patient were grown in a culture outside the body. A functional ADA cDNA, carried by a retroviral vector, was introduced into these lymphocytes, which were subsequently returned to the patient.

The limitation, and the ideal. As these cells are not immortal, the patient requires periodic infusion of such genetically engineered lymphocytes. However, if the gene isolated from marrow cells producing ADA is introduced into cells at early embryonic stages, it could be a permanent cure.

One point specific to human genes in bacteria. A human gene contains introns, which a bacterium cannot remove, so what is cloned is normally the cDNA prepared from the mature mRNA rather than the genomic sequence. This is also why the ADA gene therapy used a functional ADA cDNA.

The chemistry first. Seed oil is stored as triacylglycerol — three long-chain fatty acids esterified to one molecule of glycerol. It is non-polar and hydrophobic, which is why it is conventionally removed by mechanical pressing or by extraction into a non-polar solvent such as hexane. But both of those act on the harvested seed, and both are expensive, wasteful and leave solvent residues.

The rDNA approach: stop the oil being made at all. Triacylglycerol synthesis is an enzymatic pathway, so the way to prevent oil accumulation is to silence the gene for a key enzyme of that pathway — for instance the acyltransferase that adds the final fatty acid to the glycerol backbone.

1. Identify the target gene. Select the gene coding for the key enzyme of triacylglycerol biosynthesis in the seed.
2. Build the construct. Introduce the DNA in such a way that the host cells produce both sense and anti-sense RNA. Being complementary to each other, these form double-stranded RNA, which initiates RNA interference and silences the specific mRNA — exactly the strategy used against the nematode Meloidegyne incognitia.
3. Make it seed-specific. Place the construct under a seed-specific promoter, so that only the developing seed is affected and the rest of the plant develops normally.
4. Transform the plant. Use Agrobacterium vectors — the disarmed Ti plasmid — to deliver the construct into the plant.
5. Select and confirm. Select transformants using the marker on the vector, then confirm low oil content by chemical analysis of the seeds of successive generations.

A second rDNA route. Instead of silencing synthesis, express a lipase in the seed under a promoter active at maturation, so that the triacylglycerol already formed is hydrolysed to glycerol and free fatty acids, which are far more easily washed out or metabolised than intact oil.

Why this is better than solvent extraction. The oil is never deposited, so nothing has to be extracted: there is no solvent, no residue, no pressing loss, and the carbon the plant would have spent on oil is available for protein or starch instead. Note that this is the same logic as every application in this chapter — change the plant rather than treat the harvest.

What it is. Golden rice is a genetically modified rice enriched in vitamin A. The chapter cites it as the example of genetic modification that has enhanced the nutritional value of food.

How it works. Ordinary rice grain contains no β-carotene, the precursor from which the human body makes vitamin A, even though the rice plant makes β-carotene in its green tissue. Genes for the missing steps of the β-carotene biosynthetic pathway were introduced so that the pathway operates in the endosperm — the part of the grain that is actually eaten. The accumulated β-carotene gives the grain its golden colour, which is where the name comes from.

Why rice was chosen. Vitamin A deficiency is a leading cause of preventable childhood blindness and of reduced resistance to infection, and it is concentrated in populations for whom rice is the staple. Fortifying the food people already grow and eat every day reaches them more reliably than a supplement programme.

Where it sits in this chapter. Notice that it is a different kind of GM crop from Bt cotton. Bt cotton is modified for what it does in the field — resisting a pest; golden rice is modified for what it does for the person eating it. And like every GM crop, its release is subject to regulatory approval, in India through the GEAC.

Yes — blood contains both proteases and nucleases.

Proteases in blood. They are not digestive proteases, but they are certainly present and active.

• The blood clotting cascade is a chain of proteases: each clotting factor is a protease that activates the next by cleaving it, ending with thrombin, which cleaves fibrinogen to fibrin.
• Plasmin is the protease that dissolves clots, which is why tissue plasminogen activator is used as a recombinant therapeutic in heart attack and stroke.
• The complement system of immune defence works by a series of protease cleavages.
• Plasma also carries anti-proteases, such as α-1-antitrypsin, whose job is to keep these proteases in check — which is itself evidence that proteases are present.

Nucleases in blood. Serum contains deoxyribonucleases and ribonucleases. Their function is to degrade nucleic acid released into the circulation from dead and dying cells and from invading organisms, which prevents both the accumulation of cell debris and the uptake of foreign DNA.

Why the question is asked here. It explains a practical limit on biotechnology. A protein pharmaceutical injected into the blood will be attacked by plasma proteases, and a DNA-based therapeutic will be attacked by serum nucleases — which is why gene therapy uses a viral vector to carry and protect the DNA rather than injecting naked DNA, and why protein drugs need careful formulation and repeated dosing.

The major problem: the digestive tract is designed to destroy proteins. A protein taken by mouth meets the strongly acidic stomach, which denatures it, and then the proteases of the stomach and intestine, which hydrolyse it to its constituent amino acids. Whatever survives then faces a second obstacle: a large, hydrophilic protein crosses the intestinal epithelium very poorly. So there are really two problems, and both must be solved together — survival in the gut and absorption across the gut wall.

This is precisely why insulin has to be injected, and why almost every recombinant therapeutic protein — growth hormone, clotting factors, erythropoietin — is given by injection.

Approaches that are used or under development.

• Enteric coating. An acid-resistant coating that does not dissolve in the stomach but releases the protein in the intestine.
• Encapsulation. Enclosing the protein in liposomes, polymer nanoparticles or microspheres that shield it from proteases and release it slowly.
• Protease inhibitors. Co-administering an inhibitor so that the gut's own enzymes are temporarily suppressed.
• Chemical modification. Altering the protein — for example by attaching polyethylene glycol, or by using modified amino acids — so that proteases no longer recognise it while its biological activity is retained.
• Absorption enhancers. Agents that transiently increase the permeability of the intestinal epithelium.
• Avoiding the gut altogether. Nasal, inhaled, buccal or transdermal delivery, which bypass the digestive tract.

The trade-off to note. Every one of these solutions costs something — a lower and more variable dose reaching the blood, a more complex formulation, or a risk of letting other substances across the gut wall alongside the drug. That is why, decades after recombinant insulin was first made, it is still usually injected. Making the protein was the easy part; delivering it remains the hard one.