10.1 Containment matched to hazard #
10.1.1 Biosafety levels as an idea #
| Level | Hazard the containment is meant to match |
|---|---|
| BSL-1 | Not known to cause disease in healthy adults |
| BSL-2 | Moderate disease; prevention or treatment usually exists |
| BSL-3 | Serious or lethal disease, especially by inhalation |
| BSL-4 | Life-threatening disease, often untreated, high community risk |
The level belongs to the procedure. Adding virulence, host range, or drug resistance can raise it. Risk group is the agent; biosafety level is the work. Air handling, protective equipment, and waste treatment are laboratory training, not this text. NIH guidelines and institutional committees, and other countries’ equivalents, are the institutional heirs of Asilomar. Local approval, national law, and ethical acceptability are three gates.
10.1.2 Hazard, exposure, and the difference between a tank and a field #
Hazard is the harm. Exposure is who meets it, how, and how often. Risk is the combination. A dangerous organism held in intact containment is not the same risk as a milder one leaving the building, and a mild organism at continental scale can be the larger ecological exposure.
A fermenter succeeds when the organism stays in and the product leaves. A nutrient dependency the outdoors cannot supply is meaningful there. A crop is supposed to grow outside, so that dependency would destroy the product. Release is judged as gene flow, non-target effects, persistence, and resistance. The Cartagena Protocol of 2000 addresses transboundary movement of living modified organisms and uses precautionary language. It is not an insulin-plant rulebook. The United States is not a party. Implementation is not a single statute to memorise from these notes.
10.2 People, descendants, and access #
10.2.1 Consent for somatic gene therapy #
Consent covers purpose, procedure, risks, alternatives, and withdrawal, in understandable language, without coercion. Vector risks include inflammation, insertional cancer, immune loss of the corrected cells, and off-target edits. Those are ways the mechanism fails, not footnotes to an otherwise familiar drug.
The therapeutic misconception treats an early trial as treatment offered because it will help the subject. Seriously ill people, and parents of seriously ill children, are under pressure that makes that confusion likely. Jesse Gelsinger’s death followed adenoviral inflammation; investigation found incomplete disclosure of animal toxicity and of financial interests. A safer vector would not retire the duty to say what is known.
10.2.2 Germline editing and the case that broke the norm #
By 2018 the professional norm was that clinical human germline editing should wait for safety, a real unmet need, and broad social agreement. The norm was not a statute in every country. It was the sort of professional pause Asilomar had tried to be: a limit set by the people who understood the tool, before a birth made the limit moot.
He Jiankui announced twins whose embryos had been edited at CCR5, nominally to resist HIV. The edit is germline because it is in the embryo and heritable. Paternal transmission of HIV was already preventable, so this was not a last resort against an otherwise untreatable disease. Mosaicism and an incomplete off-target account were expected failure modes, not surprises. Oversight and consent failed. A Chinese court in 2019 convicted him of illegal medical practice. A working method is not a permission. The case does not condemn somatic therapy for a lethal immunodeficiency, and it does not close the separate argument about whether any human germline edit should ever be done.
10.2.3 Equity of access #
Insulin’s manufacturing problem can be solved while people still cannot buy it. Some gene therapies are priced beyond what many health systems will pay, and are simply unavailable in most of the world. Golden Rice meets its stated aim only if people with vitamin A deficiency actually receive it.
These are not one injustice. Insulin is a manufactured hormone with an access problem. Some gene therapies are scarce clinical procedures. A nutritional crop can fail as public health even after the pathway works, whether the barrier is regulation, politics, seed supply, or diet. Counting that gap is not an argument for skipping regulation. Risk and benefit often fall on different people: a trial subject, a neighbouring farmer, a descendant, a patient in a capital city.
10.3 Misuse, ownership, and two ways of deciding #
10.3.1 Dual-use research of concern #
Dual-use research of concern would make a pathogen more transmissible, more virulent, broader in host range, resistant to vaccine or treatment, harder to detect, or possible to reconstitute after eradication — or would provide the design for someone else to do so. A vaccine antigen and a virulence enhancement are not in that category to the same degree. The category is not a synonym for recombinant DNA.
Recognise it before the experiment. Do not perform it as ordinary work, do not split it so that nobody sees the whole, and do not treat a sequence design as harmless because it is only a file. Refer it to institutional and, where they exist, national dual-use reviews. Some experiments should not be done at all. These notes will not describe them or how published harmful work was carried out. Reviewed research on pathogens, in institutions built for it, exists because vaccines and treatments require knowledge of the organisms. That fact is not a general permission.
10.3.2 What a patent does and does not do #
A patent excludes others, for about twenty years from filing, in the country that granted it, in exchange for disclosure. It is not ownership of a wild species and not permission to ignore safety law. A patent on an engineered bacterium does not authorise release. A patent on a drug does not authorise sale before a medicines regulator agrees.
Diamond v. Chakrabarty (1980) held a human-altered bacterium patent-eligible in the United States. Myriad (2013) held that an isolated natural gene sequence is not, while cDNA can be. Other countries do not copy those two holdings. European law has its own rules, including moral exclusions of human cloning and human germline modification. A patent on seed can restrict replanting; that is an argument about control of the next planting, not about whether T-DNA works. Research exemptions differ by country. Bayh–Dole let United States universities patent publicly funded inventions. Licensing can move a strain out of a notebook, and it can also keep a trait from the people who were its stated point.
10.3.3 Precaution and risk–benefit, as a real disagreement #
The Rio formulation of precaution says that threats of serious or irreversible damage are not postponed for lack of full certainty. Stronger versions make the proposer prove sufficient safety first. Risk–benefit compares estimated harms and benefits, including the harm of the status quo. They disagree about who carries the burden of proof and about whether delay counts as harm before a mechanism is shown. More data do not always decide that.
On Golden Rice, precaution waits for irreversibility and for proof of nutritional effect in real diets. Risk–benefit counts ongoing vitamin A deficiency as harm and treats β-carotene in rice as a small biochemical risk beside it. On X-SCID, precaution stresses an integrating vector whose cancer rate was not yet known; risk–benefit stresses death from infection and accepts a stated insertional risk if families are told. The trials produced both the benefit and the LMO2 harm. Clinical germline editing in 2018 satisfied neither framework: an alternative existed, safety was inadequate for a heritable change, and reversal was at its hardest.
10.4 Key terms, summary, and self-check #
Key terms. Biosafety level; hazard; exposure; Cartagena Protocol; informed consent; therapeutic misconception; germline editing; equity; dual-use research of concern; patent; Chakrabarty; Myriad; precautionary principle; risk–benefit analysis.
Summary. Biosafety levels match containment to a procedure. Risk is hazard combined with exposure. A tank is built to hold the organism; a field is built to release it. Consent includes vector risk. He Jiankui’s embryo edits were germline, poorly justified, and sanctioned. Equity asks who can obtain the product. Dual-use research that would worsen a pathogen or toxin is recognised and not performed. A patent is a temporary right to exclude, not a release permit. Precaution and risk–benefit are a genuine disagreement about uncertainty, visible in Golden Rice and in somatic gene therapy, and neither excused the 2018 germline case.
Self-check.
1. Why can one species be handled at more than one biosafety level?
2. Why is a nutritional dependency sensible containment for a laboratory bacterium and a bad design for a crop?
3. What made the He Jiankui edits germline, and why was the HIV rationale inadequate?
4. What does a granted patent on a transgenic microbe not decide?
5. For Golden Rice or the X-SCID trials, name one weight precaution gives that a bare risk–benefit sum may discount, or the reverse.
—
# Closing: how the chapters fit together
The object of the notes is a defined DNA change. Chapter 1 separates it from a selected variant and records the 1975 containment argument. Chapters 2 to 4 build, carry, express, and prove the molecule. Later work fails when that proof is skipped: an unsequenced edit, an unbalanced pathway, and an unmapped plant insertion are the same error as an untested plasmid map.
Chapter 5 moves the target onto a chromosome. Chapter 6 moves it onto a protein or a flux, and refuses to call artemisinic acid a factory. Chapter 7 is the factory, with insulin as the medicine that made recombinant DNA ordinary. Chapters 8 and 9 leave the tank. Plants add gene flow, resistance, and a legal split between insertion and a small edit that is not the same split everywhere. Medicine adds the vector as a drug, the LMO2 lesson, CAR-T as a receptor, and nuclear transfer as copying. Chapter 10 is the judgement none of those techniques can make for themselves.
Cre–lox links conditional mice to marker removal. CHO glycosylation links Chapter 3 to the medicines of Chapter 9. Clarke–Carbon arithmetic is why BACs exist. A Bt refuge, antibiotic resistance, and a chemostat mutant are one selection argument. Biological containment suits a disabled production strain and contradicts a crop bred to survive outdoors.
Revision checklist #
1. Separate a selected strain from a defined construct, and state Asilomar’s two containments.
2. Give the spacing of a six-base site, sticky versus blunt ends, and one reason a mapped site does not cut.
3. State what PCR copies, what proofreading changes, and why cDNA lacks promoters.
4. Order plasmid, lambda replacement, cosmid, BAC, and YAC by insert, and say why therapeutic antibodies are made in CHO cells.
5. Give two blue–white failures, and say what whole-plasmid sequencing adds to a digest.
6. Distinguish NHEJ, HDR, base editing, and dead Cas, and somatic from germline editing.
7. Write the directed-evolution cycle, and say why artemisinic acid in yeast was not yet a supply.
8. Compare batch, fed-batch, and chemostat, including washout, and say why recombinant insulin replaced animal insulin.
9. Trace T-DNA from wound signal to disarmed insert. Name a Bt assumption and what Golden Rice accumulates.
10. Define risk. State what a patent does not authorise, what dual-use concern requires of you, and one precaution versus risk–benefit disagreement.
Selected sources #
These notes are an original teaching summary, not a transcript of any textbook.
- Brown TA. Gene Cloning and DNA Analysis: An Introduction. Wiley-Blackwell.
- Primrose SB, Twyman RM. Principles of Gene Manipulation and Genomics. Blackwell / Wiley.
- Glick BR, Patten CL. Molecular Biotechnology: Principles and Applications of Recombinant DNA. ASM Press. Earlier editions by Glick and Pasternak are the same tradition.
- Nicholl DST. An Introduction to Genetic Engineering. Cambridge University Press.
- Cohen SN, Chang ACY, Boyer HW, Helling RB. Construction of biologically functional bacterial plasmids in vitro. Proceedings of the National Academy of Sciences USA 70:3240–3244 (1973).
- Berg P, Baltimore D, Brenner S, Roblin RO, Singer MF. Summary statement of the Asilomar conference on recombinant DNA molecules. Science 188:991–994 (1975).
- Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337:816–821 (2012).
- Ro DK, Paradise EM, Ouellet M, et al. Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature 440:940–943 (2006).
- Ye X, Al-Babili S, Klöti A, Zhang J, Lucca P, Beyer P, Potrykus I. Engineering the provitamin A (β-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 287:303–305 (2000).
- Wilmut I, Schnieke AE, McWhir J, Kind AJ, Campbell KHS. Viable offspring derived from fetal and adult mammalian cells. Nature 385:810–813 (1997).
—
—
# Part C. Visual Figures & Scientific Diagrams Gallery
Comprehensive high-resolution visual reference repository for Biotechnology & Genetic Engineering from the B10logy Visual Atlas.
Gel Electrophoresis In Dna Fingerprinting
—
# Part C. Visual Figures & Scientific Diagrams Gallery
Essential high-resolution reference diagrams for Biotechnology and Genetic Engineering from the B10logy Visual Atlas.