Synthetic Biology and Bioengineering

A Tier 2 specialty covering the engineering discipline that treats biology as a programmable substrate. Synthetic biology arose at the intersection of molecular biology, computer science, and chemical engineering circa 2000 with the publication of three founding papers in Nature: Elowitz-Leibler’s repressilator oscillator, Gardner-Cantor-Collins’s genetic toggle switch, and the BBa/MIT BioBrick parts registry concept. By 2026 the field has matured from one-off academic devices into a $20B industrial pipeline spanning therapeutics (CAR-T, gene therapy, mRNA), industrial enzymes, novel materials (spider silk, mycelium), agriculture (Pivot Bio N-fixation), and consumer products (Impossible heme, Perfect Day whey, Brewed Protein silk-replacement). This note covers the design-build-test-learn engineering cycle, DNA assembly methods, genetic circuit design, the major chassis organisms, metabolic engineering principles, cell-free systems, biosensors, minimal cells, and industrial-scale considerations.

See also

• cell-molecular-biology
• genetics-and-genomics
• microbiology-foundations
• structural-biology
• virology-and-vaccine-platforms
• proteomics-metabolomics-and-computational-neuroscience
• model-organisms-and-sequencing-tech
• protein-families-and-drug-targets

---

The design-build-test-learn (DBTL) cycle

Synthetic biology projects iterate around a four-step engineering loop borrowed from electrical and chemical engineering:

1. Design. Specify the desired phenotype (titer of small molecule X; circuit topology Y; biosensor response Z). Select chassis, parts (promoters, RBSs, CDSs, terminators), DNA-level design tool (Benchling, Geneious, SnapGene, j5/Teselagen for combinatorial design, Cello for circuit compilation). Submit DNA orders to vendor (Twist, IDT, GenScript, Ginkgo Foundry custom).
2. Build. Assemble DNA constructs (Gibson, Golden Gate, MoClo, BioBrick); transform into chassis; pick colonies; verify by colony PCR + Sanger sequencing (or Nanopore plasmid sequencing as the new default). For library construction, transform in 96/384-well plates with automation (Opentrons OT-2, Hamilton STAR, Beckman Echo).
3. Test. Phenotype: growth curves, fluorescence kinetics (Tecan Spark, BioTek Synergy, BMG ClarioStar; flow cytometry); LC-MS/GC-MS for metabolite quantification; RNA-seq / proteomics for systems-level diagnostics.
4. Learn. Statistical models, mechanistic kinetic models, ML (active-learning, Bayesian optimization, generative protein design). Update part library, codon tables, FBA fluxes.

Cycle time has compressed from ~12 months in 2005 (hand-cloning, gel-purified fragments) to ~1-3 weeks in 2026 at well-funded labs (automated cloning + sequencing + screening). Ginkgo Bioworks, Inscripta, Lygos, Zymergen-acquired-by-Ginkgo, Amyris, Calysta, Solugen all run industrial-scale DBTL platforms.

---

DNA synthesis and assembly

DNA synthesis

• Oligo synthesis (~10-200 nt). Phosphoramidite chemistry (Caruthers 1981). Yield ~99.5%/cycle → ~37% intact at 200 nt without correction. Major vendors: IDT (Integrated DNA Technologies), Eurofins MWG, Sigma-Aldrich (Merck) — turnkey within days for standard ssDNA, dsDNA, primers.
• Gene synthesis (~100 bp - 10 kb). Oligo assembly + error correction. Vendors: Twist Biosciences (silicon-based 10k-oligo chips, low cost — ~US$0.07/bp2024),IDTgBlocks(US$0.09/bp), GenScript, Eurofins. Twist dominated 2018-2024 on price; SDI, Ansa Biotechnologies, and emerging enzymatic-synthesis startups challenge.
• Enzymatic DNA synthesis. TdT (terminal deoxynucleotidyl transferase) with reversibly blocked nucleotides. Avoids harsh chemistry, longer reads, lower error. DNA Script SYNTAX (1st commercial benchtop 2021), Molecular Assemblies, Camena Bioscience, Ansa Biotechnologies — multiple competing pipelines.

Gibson assembly

Gibson-Smith-Hutchison-Venter 2009 Nat Methods. Single-pot, isothermal (50 °C, 1 h) assembly. Three enzymes:

1. T5 exonuclease chews back 5’ ends.
2. Phusion polymerase fills gaps.
3. Taq DNA ligase seals nicks.

Inputs: linear dsDNA fragments with ~20-40 bp overlapping homology regions at each junction. Assembles 2-15 fragments simultaneously, fragment lengths from 100 bp to 100 kb. Used at JCVI to assemble the M. mycoides JCVI-syn1.0 genome from synthesized oligos (Gibson 2010 Science). Commercially: NEB Gibson Assembly Master Mix, NEBuilder HiFi.

Golden Gate / MoClo

Golden Gate (Engler-Marillonnet 2008) uses Type IIS restriction enzymes (BsaI, BsmBI/Esp3I, BbsI) that cut outside their recognition site, leaving programmable 4-nt overhangs. Recognition sites are eliminated upon ligation → single-pot reaction can do digestion + ligation simultaneously with thermocycling 37/16 °C (or 37 °C only with high-fidelity rebranded enzymes — NEB BsaI-HFv2, Esp3I).

MoClo (Modular Cloning) — Werner-Engler-Marillonnet 2011 hierarchical scheme:

• Level 0 — basic parts (promoter, 5’UTR, CDS, 3’UTR, terminator) with standardized BsaI overhangs.
• Level 1 — transcription units (promoter + CDS + terminator) from Level 0 parts.
• Level 2 — multigene constructs (3-7 TUs) from Level 1 transcription units.
• Level M / Level P — toolkit-specific super-constructs.

MoClo toolkits: original plant Engler toolkit, MoClo Yeast Toolkit (YTK, Lee-Cai-Engler-Dueber 2015), iGEM Phytobrick standard, Bacterial Toolkit (CIDAR, Densmore lab), MammoBlock, MarMo for cyanobacteria.

BioBrick

Knight 2003. RFC10 standard: prefix-suffix flanking restriction sites (EcoRI, NotI, XbaI for prefix; SpeI, NotI, PstI for suffix). Idempotent assembly. iGEM Registry of Standard Biological Parts catalogs ~20k parts. Now largely supplanted by Gibson and Golden Gate in research labs because BioBrick’s “scarred” 6-8 bp junction between every assembled part interferes with biology (can change ribosome binding, introduce stop codons in fusion proteins). BioBrick remains a teaching standard.

Other assembly methods

• CPEC — circular polymerase extension cloning (Quan-Tian 2011); no enzymes beyond polymerase; PCR-only.
• In-Fusion (Takara) — 15-bp overlap; proprietary enzyme.
• SLIC, LIC, SLiCE — sequence/ligation-independent cloning variants.
• AQUA cloning — quick assembly using yeast recombination.
• Yeast transformation-associated recombination (TAR). S. cerevisiae homologous recombination assembles megabase-scale constructs from overlapping fragments — used at JCVI for synthetic mycoplasma genome and for chromosome-scale designs.

Codon optimization

For heterologous expression, codon usage of the host matters. CAI (codon adaptation index), CAI table per organism. Tools: IDT Codon Optimization Tool, Twist Codon Optimization, GeneArt (Thermo), JCat. Caveats: extreme codon optimization can reduce protein yield by removing translation-pause codons that aid co-translational folding; mRNA secondary structure around the start codon matters more than CAI; experimentally measured “optimal” codons differ from CAI for some hosts.

---

Genetic circuits

Standard parts

• Promoters. Constitutive (E. coli J23xxx Anderson series, T7, P_BAD, lac, tac, trc), inducible (PLtetO1/tet, lacUV5/IPTG, P_BAD/arabinose, pCu/copper, NahR/salicylate), repressible (tetR, lacI). Yeast: TEF1, PGK1, TDH3, GAL1 (galactose-inducible), CUP1 (Cu-inducible). Plants: 35S CaMV, ubiquitin, RUBISCO. Mammalian: CMV, EF1α, CAG, SV40, doxycycline-inducible (tet-on/off), cumate-inducible, blue-light-inducible.
• RBS (E. coli) / Kozak (eukaryotes). Salis RBS Calculator (Salis 2009 Nat Biotechnol) predicts translation initiation rate from sequence → allows tuning protein expression ~10⁴-fold by RBS choice alone.
• Terminators. Rho-independent (intrinsic) for prokaryotes; PolyA for mammalian; native gene terminators (CYC1, ADH1) for yeast.
• Insulators / spacer sequences. Reduce context-dependence at junctions.
• Ribozymes (RiboJ). Lou-Stanton-Voigt 2012 — insulates RBS from upstream sequence.

Toggle switch (Gardner-Cantor-Collins 2000)

Two repressors mutually repressing each other (lacI and cI) + inducers (IPTG, anhydrotetracycline). Bistable; flips between two stable states with transient inducer pulse. First synthetic genetic memory device.

Repressilator (Elowitz-Leibler 2000)

Three transcriptional repressors in a cyclic loop (lacI → tetR → cI → lacI). With sufficient cooperativity (Hill coefficient n ≥ 2) and ratio of degradation rates, system oscillates with period ~150 min in E. coli. Original repressilator showed wide cell-to-cell phase variability; redesigned versions (Potvin-Trottier-Lord-Vinnicombe-Paulsson 2016 Nature) used lower transcriptional noise and synchronized oscillations across populations.

Logic gates

Boolean AND, OR, NAND, NOR, XOR implemented at:

• DNA layer. Recombinase-based (FlpO, Cre, Bxb1, PhiC31) flip orientation to enable/disable downstream expression.
• Transcription layer. Hybrid promoters combining multiple operators (Hybrid pTet/pLac = AND of tet + lac inducer absence; Voigt lab).
• Translation layer. Toehold switches (Green-Silver-Collins 2014 Cell) — RNA hairpin sequesters RBS until complementary trigger RNA opens it.
• Protein layer. Split proteins (split-GFP, split-Cas9, split-T7-RNAP) — fragments interact only when fused to interaction domains.
• Cell layer. Two-cell consortia computing distributed functions.

Cello (Nielsen-Voigt 2016 Science) — first SBOL-based EDA tool for synthetic circuits; user specifies Verilog-like logic, Cello compiles to DNA layout with characterized parts from a library. Predicted-vs-measured circuit fidelity ~70% functional on first build.

Quorum sensing and communication

LuxI/LuxR (Vibrio fischeri; AHL signaling), LasI/LasR (Pseudomonas), agr (S. aureus). Engineered consortia: predator-prey, divide-of-labor microbial communities. Pattern formation in spatially structured biofilms (Liu-Süel 2015 Nature — electrochemical biofilm signaling).

Genetic memory

• Recombinase-based memory. Bxb1, PhiC31 (Bonnet-Endy-Subsoontorn-Endy 2012, 2013) — DNA inversions are permanent (no protein-decay required). Programmable to 1-bit, 2-bit, … states. Wong-Liu 2016 Cell “Record-keeper”: stores up to 1.4 trillion bits in DNA across a population.
• CRISPR-Cas-based memory. Type I-E CAS (Shipman-Nivala-Macklis-Church 2017 Nature) — encodes information in CRISPR arrays by spacer acquisition.
• DNA storage. Synthesizing arbitrary digital information into oligo libraries; Microsoft-UW + Twist commercial DNA data storage. ~200 PB/g theoretical density.

---

CRISPR for synthetic biology

Cross-link genetics-and-genomics for editing fundamentals.

CRISPRi / CRISPRa

• CRISPRi. dCas9 (catalytically dead Cas9 D10A H840A) + sgRNA → sequence-specific transcriptional repression (Qi-Lim-Weissman 2013 Cell). Genome-scale CRISPRi screens in mammalian cells (Gilbert-Larson-Doudna-Weissman 2013, 2014).
• CRISPRa. dCas9 fused to VP64, p65, Rta, or SAM (synergistic activation mediator: dCas9-VP64 + MS2-p65-HSF1) → transcriptional activation.

CRISPRi/a libraries (CRISPRi-v2, hCRISPRi-v2 from Doench, CRISPRko/i/a packs from Brunello/Calabrese) are now standard for functional genomics.

Base editors

• CBE (cytidine base editor). Komor-Liu 2016 Nature. APOBEC1 (rat) or AID + nCas9 (D10A nickase) + UGI (uracil glycosylase inhibitor) → C·G → T·A. BE3, BE4max, evoBE4max (Hyperthermus butylicus).
• ABE (adenine base editor). Gaudelli-Liu 2017 Nature. Evolved E. coli TadA tRNA adenosine deaminase + nCas9 → A·T → G·C. ABE7.10, ABE8e (Richter-Liu 2020 highly active).

Avoids dsDNA breaks → reduced off-target indels; very precise for SNV correction. Verve Therapeutics PCSK9 ABE therapy (VERVE-101) — first base-edited drug in clinic 2022; PCSK9 inhibition for cardiovascular disease.

Prime editors

Anzalone-Liu 2019 Nature. nCas9 + reverse-transcriptase (M-MLV variant) fused; pegRNA encodes both target site AND desired edit. Up to 20-nt insertions, all 12 base substitutions, small deletions. PE2, PE3, PE3b, PEmax. Lower efficiency than base editing but unrestricted edit types. Prime Medicine clinical program.

Other CRISPR variants

• Cas12a (Cpf1). Zhang lab. T-rich PAM (TTTV); staggered cuts; smaller than Cas9.
• Cas13 (Abudayyeh-Gootenberg-Zhang). Targets RNA. Used in SHERLOCK diagnostics.
• CasX, Cas14, Cas12f. Smaller; expanding capability.
• CasΦ. Discovered in phage genomes; minimal size.

---

Cell-free systems

In vitro protein synthesis without intact cells. Two main paradigms:

Lysate-based (TX-TL, CFPS, S30, S12)

Cell extracts retain ribosomes, tRNAs, RNA polymerase, translation factors. Supplement with energy regeneration (PEP/PK, 3-PGA/PK, creatine phosphate/CK, or maltose/maltodextrin), nucleotides, amino acids, plasmid or linear DNA template.

• PURExpress (NEB; Shimizu-Ueda PURE system) — fully reconstituted from 36 purified components; cleanest for biophysics; expensive.
• myTXTL (Arbor Biosciences) — E. coli extract; commercial CFPS.
• CFAB — cell-free active biocatalysis.

Applications: biosensor prototyping, paper-based diagnostics (Pardee-Collins 2014 Cell Ebola/Zika test), enzyme discovery, on-demand drug manufacturing (Pardee 2016 Cell), education and outreach.

Continuous-flow CFPS

Spirin’s continuous-exchange cell-free (CECF) for long-term synthesis. Now Sutro Biopharma platform (Xpress CF) — commercial protein production of difficult ADCs and bispecifics.

Glucose-fed cell-free (Bowie lab)

Replace ATP regeneration with glucose oxidation in a controlled chassis-free metabolism — extends reaction lifetime to >24 h.

---

Chassis organisms

Escherichia coli

Workhorse for recombinant protein, plasmid prep, cloning, prototyping. Strains:

• K-12 lineages. MG1655 (sequenced reference), DH10B/MachT1 (cloning), DH5α (cloning), BL21(DE3) and derivatives (Rosetta — rare codons; Origami — disulfide-friendly; Lemo21(DE3) — tunable T7 RNAP).
• BL21 derivatives. lon-deficient, ompT-deficient → reduced protease activity; ideal for protein production.
• Nissle 1917. Probiotic strain; gut chassis for therapeutic engineering (Synlogic).
• Genome-reduced. E. coli MDS42 (Pósfai-Plunkett 2006 Science — 14% genome reduction); GR-Ec series.

Limitations: no glycosylation (workaround: ER-mimic in vitro plus PglB N-glycosylation pathway transplanted from Campylobacter; Aebi-DeLisa); endotoxin (LPS — must be removed for therapeutic protein); secretion only via Sec/Tat/T1SS engineering; codon bias for rare tRNAs.

Bacillus subtilis

Gram-positive; natural competence; native secretion (no LPS); GRAS. Industrial enzymes (Novozymes, Genencor/IFF, Sumitomo) — proteases, amylases, lipases, cellulases secreted at 10-30 g/L. Cross-link microbiology-foundations.

Saccharomyces cerevisiae

Best-studied eukaryote. Native HR enables CRISPR efficiency ~100% in single transformation. Plasmids: CEN/ARS (low copy), 2µ (high copy), integrative (YIP). Strains: BY4741, BY4742 (deletion library parents), CEN.PK113-7D (industrial), W303-1A.

Industrial: bioethanol (Brazil sugarcane, US corn — ~120 Mt/yr globally), beer/wine, recombinant insulin (Novo Nordisk), hepatitis B vaccine (GSK Engerix, MSD Recombivax), HPV vaccine (MSD Gardasil — VLPs from S. cerevisiae), human serum albumin (Albagen, Mitsubishi Tanabe).

Pichia pastoris (Komagataella phaffii)

Methylotrophic yeast; AOX1 methanol-inducible promoter; very high secreted protein titers (5-15 g/L in fed-batch). GlycoSwitch engineered humanized N-glycan (GlycoFi, acquired by Merck). Companies: Pfenex (now Ligand), Phyton, Mitsubishi, Sumitomo.

CHO cells

Chinese Hamster Ovary; dominant therapeutic mAb host (~80% of marketed mAbs). CHO-K1, CHO-DG44 (DHFR-), CHO-S, CHO-GS (glutamine synthetase). Titers in fed-batch 5-12 g/L; perfusion >25 g/L. Glycosylation similar to human but with non-human Gal-α-1,3-Gal and Neu5Gc — strain engineering (Lonza GS, Horizon GS-KO, Wuxi Biologics, Boehringer Ingelheim) to humanize.

Other chassis

• HEK293, HEK293T, HEK293F. Transient transfection; gene therapy vector production (AAV, lentivirus).
• Sf9, Sf21, High Five. Baculovirus expression for difficult eukaryotic proteins.
• Streptomyces coelicolor, Streptomyces avermitilis. Native polyketide/NRPS producers for antibiotic discovery and engineering.
• Yarrowia lipolytica. Oleaginous yeast; lipid biofactory (Conagen, Calysta-related single-cell protein from methanol).
• Cyanobacteria Synechocystis sp. PCC 6803, Synechococcus elongatus PCC 7942 — phototrophic chemicals.
• Chlamydomonas reinhardtii — algal chassis.
• Vibrio natriegens — fastest-doubling bacterium (~10 min); emerging fast cloning host.
• Mammalian cell platforms — Jurkat for T-cell engineering, HepG2 for liver, iPSC.

---

Metabolic engineering

The branch of synthetic biology aimed at producing valuable small molecules (chemicals, fuels, drugs) by engineering microbial metabolism.

Flux balance analysis (FBA)

Genome-scale metabolic models (GEMs): iJO1366 (E. coli; Orth-Conrad-Palsson 2011), iMM904 (S. cerevisiae), Recon3D (human, Brunk-Palsson). Stoichiometric matrix S (m metabolites × n reactions); steady state Sv = 0; bounds α ≤ v ≤ β; objective max c·v (typically biomass or product formation).

Solved by linear programming. Tools: COBRA Toolbox (MATLAB, Python cobrapy), Optflux, Cameo, Memote (QC), KEGG/BiGG/MetaCyc databases.

13C metabolic flux analysis (13C-MFA)

Feed isotopically labeled substrate (typically [1-13C]glucose, [U-13C]glucose, or mixtures), measure 13C enrichment patterns in proteinogenic amino acids by GC-MS / LC-MS / NMR. Software: INCA (Antoniewicz), 13CFLUX2, OpenFLUX, FluxPyt. Resolves split ratios at branch points (G6P → PPP vs glycolysis; PEP → OAA via PEPC vs PYR via PK).

Strain construction strategies

• Knockout undesired flux drains. Lactate dehydrogenase deletion to redirect pyruvate → product.
• Overexpress bottleneck enzymes. Stronger promoter, multiple copies, codon optimization.
• Heterologous pathway transplant. Move metabolic pathway from native (often unculturable) to chassis. E. coli synthesizing artemisinic acid (Amyris/Sanofi semi-synthetic artemisinin — replaced 60% of global supply from Artemisia annua plant), opioids (Smolke 2015 Science opiate biosynthesis from glucose in yeast), cannabinoid biosynthesis (Keasling 2019).
• Cofactor balancing. NADH vs NADPH; ATP yield; redox cofactor swapping with engineered ene-reductases.
• Transport engineering. Substrate uptake (lacY for lactose, malEFG for maltose), product export (efflux pumps to relieve product toxicity).
• Dynamic regulation. Sense pathway intermediate, regulate upstream enzyme expression to maintain balance. Pyruvate biosensor → acetate pathway.
• Compartmentalization. Carboxysome, encapsulin, peroxisome — confine pathways to organelles; concentrate intermediates.

Industrial metabolic engineering successes

• 1,3-Propanediol (Sorona, DuPont-Tate&Lyle). E. coli engineered to ferment glucose to 1,3-PDO; precursor to PTT polyester. >100 kt/yr commercial plant Loudon, TN.
• Succinic acid (Reverdia, Myriant, BioAmber). Engineered E. coli, yeast (Saccharomyces, Yarrowia); precursor to bio-based plastics, solvents. Commercial 2010-2015; market crashed when shale gas dropped petrochemical succinic price.
• 1,4-Butanediol (Genomatica BDO). Engineered E. coli; ~30 kt/yr Cargill facility (Eddyville, IA); displaced petrochemical route.
• Artemisinic acid. Amyris-Sanofi; cited above.
• Farnesene (Amyris). Yeast engineered for sesquiterpene; jet fuel precursor, fragrance, lubricants.
• Squalene, squalane (Amyris, Aprinnova). Shark-free skincare ingredient.
• Cannabinoids (Cronos-Ginkgo, Hyasynth, Demetrix, Librede). Pichia and Saccharomyces; pre-clinical to early commercial.
• Vanillin (Solvay, Evolva). Yeast vanillin from glucose or ferulic acid → vanillin.
• Lactic acid (NatureWorks). Lactobacillus; >200 kt/yr; PLA bioplastic.
• Heme (Impossible Foods). Pichia-produced leghemoglobin → meat-substitute flavor. FDA GRAS 2018.
• Animal-free milk proteins. Perfect Day (β-lactoglobulin, αs1-casein in Trichoderma); New Culture (mozzarella).
• Spider silk (Bolt Threads, Spiber). Engineered yeast or bacteria producing recombinant silk proteins for textiles.
• N-fixing bacteria for agriculture (Pivot Bio). Engineered Klebsiella, Kosakonia to fix N₂ under fertilizer-rich conditions; reduces synthetic N input on corn ~20-25 lb N/acre.

---

Biosensors

Genetically encoded sensors couple biology to readouts. Key designs:

• Transcription-factor based. TF binds analyte → activates reporter expression. E.g., AraC/araBAD (arabinose), TetR/tet (tetracycline), MerR (mercury), LuxR (AHL). Engineer for new analytes by directed evolution of binding pocket.
• Allosteric protein switches. Maltose-binding protein with inserted GFP → maltose changes fluorescence.
• FRET sensors. CFP-linker-YFP; conformational change on analyte binding → FRET ratio change. Cameleons (Ca²⁺), Camgaroo, GCaMP (single-FP Ca²⁺ sensor — backbone of all modern in vivo Ca²⁺ imaging; cross-link neuroscience-foundations).
• Single-FP intensity sensors. GCaMP6, GCaMP7, jGCaMP8 (Looger-Svoboda Janelia) — calcium; iGluSnFR (glutamate); GRAB sensors (Li-Yulong 2018-present) — dopamine, serotonin, acetylcholine, norepinephrine, ATP.
• Riboswitches. RNA aptamer-controlled gene expression; theophylline, glucosamine-6P, FMN-binding natural riboswitches.
• Whole-cell biosensors. E. coli or yeast engineered to express reporter upon environmental analyte detection. Field-deployable; paper-based.

Commercial sensors: Synlogic-Roche SYNB1934 PKU therapeutic (engineered E. coli Nissle); Microvi (water treatment); ENvueAg (agricultural pathogen). Pardee paper-based Zika sensor (2016 Cell) used cell-free with toehold switches.

---

Minimal cells

How few genes does life need? JCVI program (Venter, Smith, Hutchison) chemically synthesized a minimized Mycoplasma mycoides genome:

• JCVI-syn1.0 (Gibson 2010 Science) — synthetic copy of the 1.08 Mb M. mycoides genome; ~880 genes; first synthetic-genome cell.
• JCVI-syn3.0 (Hutchison 2016 Science) — minimized to 531 kb / 473 genes by iterative deletion. ~149 essential genes of unknown function (“quasi-essential genes”). Slow doubling (~3 h vs ~1 h native).
• JCVI-syn3A (2019) — 19 genes added back to restore normal cell morphology and faster growth; ~493 genes. Now reference minimal-cell platform.

Sc2.0 (synthetic yeast 2.0; Boeke-Chandrasekaran-Bader-Tian consortium, ~2011-present) is synthesizing a designer S. cerevisiae genome — all 16 chromosomes built and combined into a single strain by 2023; PCR-tagged loxPsym sites enable SCRaMBLE rearrangement to evolve new genotypes on demand.

---

Xenobiology

Engineering biology with non-natural building blocks.

Expanded genetic code

Schultz-Yale-Romesberg pioneered orthogonal amber-suppressor tRNA/aminoacyl-tRNA-synthetase pairs (typically from M. jannaschii tyrosyl-RS evolved against E. coli host) that load unnatural amino acids (UAAs) site-specifically at amber stop codons. Over 200 UAAs now genetically encoded: photocrosslinkers (pBpa, AzF), bioorthogonal handles (azide, alkyne, BCN, TCO), fluorophores, post-translational mimics (phospho-Ser, acetyl-Lys, methyl-Lys).

Genomically recoded organism (Lajoie-Church 2013 Science) — replaced all 321 amber codons in E. coli genome with ochre → freed amber for UAA incorporation throughout proteome; resistant to amber-suppressor phages.

Synthetic E. coli Syn61.Δ3 (Fredens-Chin 2019 Nature; Robertson-Chin 2021 Science) — E. coli genome with all TCA, TCG codons replaced by AGC, AGT (compressed serine codons), and TAG (amber) replaced by TAA → freed three codons that can be reassigned to UAAs. Resistant to T7 and most natural phages.

Xenonucleic acids (XNA)

DNA/RNA backbone analogs:

• TNA (threose). 4-membered sugar; Pinheiro-Holliger 2012 evolved TNA-binding XNA polymerases.
• HNA (hexitol), CeNA (cyclohexenyl). Pinheiro-Holliger.
• PNA (peptide nucleic acid). Nielsen 1991; charge-neutral; hybridizes DNA/RNA.
• FANA, ANA, 2’-F-RNA, LNA, BNA. Sugar-modified XNAs; some clinical (LNA, 2’-OMe in antisense oligos).

XNA aptamers and ribozymes engineered (Taylor-Holliger 2015 Nature) — synthetic genetic polymers can encode information and catalyze reactions, demonstrating non-DNA “alien” life is chemically feasible.

Hachimoji DNA (8-letter)

Hoshika-Benner 2019 Science — pairs of P-Z and B-S synthetic bases on top of A-T-G-C; eight-letter alphabet; XNAse compatible polymerase enzymes; demonstrated information encoding.

Unnatural base pair (UBP) in living E. coli

Romesberg lab — dNaM-dTPT3 hydrophobic UBP; incorporated by engineered nucleotide transporters and DNA polymerases; maintained through cell division (Zhang-Romesberg 2017 Nature). Used to genetically encode UAAs at new codons. Founded Synthorx (acquired by Sanofi 2019 for $2.5 B) for “SuperImmunes” engineered cytokines.

---

Industrial scale-up

Bioreactors

• Stirred-tank. 0.5 L lab → 200 L pilot → 25,000-200,000 L commercial. Sartorius BIOSTAT, ABEC, Pall (now Cytiva), Pierre Guerin, GEA. Aeration 0.1-2 vvm; impeller tip speed limited by shear sensitivity.
• Air-lift. No mechanical impellers; lower shear; large-scale single-cell protein (Calysta FeedKind from methane); Quorn mycoprotein.
• Membrane bioreactors. Hollow fiber for perfusion CHO; very high cell densities (>50 × 10⁶ cells/mL).
• Photobioreactors. Flat-panel (Subitec), tubular (AlgaTech), open raceway pond. Microalgae and cyanobacteria.
• Solid-state fermentation. Fungi (Trichoderma) on solid substrate; enzymes, food fermentations.

Single-use bioreactors

Stainless-steel CIP/SIP replaced by gamma-irradiated single-use bags (Sartorius Biostat STR, Pall Allegro, Cytiva Xcellerex, Thermo HyPerforma). Reduced cleaning validation, faster changeover, lower contamination risk. Now dominant for clinical and early-commercial mAb production.

Continuous bioprocessing

Perfusion + continuous capture chromatography (Genentech, Sanofi mAb continuous demos). Smaller footprint, more steady output, integrate with downstream continuous polishing.

Downstream (DSP)

• Cell removal. Centrifuge, depth filter, tangential flow filtration.
• Capture chromatography. Protein A (mAbs), IMAC (His-tag), affinity tags. Cytiva MabSelect SuRe, Tosoh, Repligen.
• Polishing. Ion exchange (cation, anion), HIC, mixed-mode, size-exclusion.
• Virus removal/inactivation. Low-pH hold, solvent/detergent, virus-removal filters (Planova, Viresolve).
• Ultrafiltration/diafiltration. Final concentration and buffer exchange.

Cost economics

mAb cost-of-goods 2024 ~US$50-200/g(dependingontiter,scale);biosimilarsoften30-50$5/kg (Novozymes scale). Bio-based fine chemicals (squalane, terpenes) US$10-100/kg.Bulkbio-chemicals(1,3-PDO,BDO)US$2-5/kg — competitive with petrochemical.

Regulatory landscape

• FDA / EMA / PMDA. GMP-grade biologics (mAbs, enzymes, gene therapy); cell-line bank characterization, host-cell protein/DNA limits, viral safety, comparability.
• FDA Coordinated Framework (US). Splits jurisdiction across FDA (therapeutics, food), USDA (livestock, plants), EPA (pesticides, microbial pest products). Inefficient but reformable under USCA.
• Cartagena Protocol. International framework for GMO trade; most engineered crops require national approval per country.

---

Practical workflows

Cloning a heterologous pathway into E. coli (Gibson)

1. Design: Choose chassis strain, plasmid backbone (pET-28a for IPTG-inducible, pBAD24 for arabinose, pSB1A2 for cloning), promoters per gene, RBS strength via Salis calculator, codon-optimize for E. coli.
2. Order gene fragments (Twist gBlocks ~$0.07/bp).
3. PCR-amplify backbone with primers carrying 20-30 bp homology to first/last fragments. Verify on gel; DpnI digest template.
4. NEBuilder HiFi mix: equimolar fragments (~0.05 pmol each), incubate 50 °C 60 min.
5. Transform 2 µL into NEB 10-beta or DH5α; plate on selective LB.
6. Colony PCR + Sanger sequencing (or Plasmidsaurus Nanopore plasmid sequencing — entire plasmid for ~$15).
7. Move into expression strain (BL21(DE3) or Rosetta(DE3) for rare codons); test expression at 25 °C overnight with 0.1-1 mM IPTG.
8. LC-MS or LC-UV phenotype assay; iterate.

CRISPR knockout in mammalian cells (lentiviral)

1. Design sgRNA via CRISPick (Doench), Benchling, or CHOPCHOP — top 3 hits with off-target score.
2. Clone into lentiCRISPRv2 (or lentiGuide-Puro + separate lenti-Cas9-blast).
3. Co-transfect HEK293T with viral plasmid + psPAX2 + pMD2.G for lentivirus packaging (Addgene). Harvest 48-72 h. Filter through 0.45 µm; concentrate by ultracentrifugation if needed.
4. Transduce target cells at MOI 0.3 (single-copy integration); select 1-2 weeks puromycin.
5. Single-cell clone (limiting dilution or FACS into 96-well); expand 2-4 weeks.
6. Genotype: TIDE/ICE Sanger analysis or NGS (CRISPResso2); western for protein loss.

---

Automation and biofoundries

The industrialization of synthetic biology has produced “biofoundries” — automated DBTL platforms that compress the cycle to days and operate at thousands-of-strains-per-week throughput. The Global Biofoundry Alliance (GBA, 2019) coordinates 30+ academic and industrial foundries.

Hardware

• Liquid handlers. Hamilton STAR/STARlet/Vantage, Tecan Fluent/Freedom EVO, Beckman Echo (acoustic dispense; nanoliter resolution; no tips), Opentrons OT-2/Flex (open-source, low-cost), Eppendorf epMotion, SPT Labtech mosquito.
• Plate handlers. PerkinElmer JANUS, Hudson PlateCrane, Yaskawa Motoman robot arms with HighRes integration.
• Incubators. Cytomat (Thermo) automated CO2 incubators with shaking; Liconic STX/SVR.
• Plate readers. Tecan Spark, BMG ClarioStar/PHERAstar, BioTek Synergy/Cytation (multi-mode + imaging).
• Sequencers. Illumina NextSeq/NovaSeq for amplicon barcode counting; Oxford Nanopore MinION/PromethION for plasmid sequencing.
• Mass spec. SCIEX 7500, Thermo Q Exactive, Waters BioAccord — high-throughput LC-MS for metabolite quant.
• Colony picking. Molecular Devices QPix, Singer Instruments PIXL.

Software / LIMS

Ginkgo Bioworks Foundry, Twist DNA design tools, Benchling, TeselaGen, Asimov Kernel, AutoProtocol (Transcriptic now Strateos), Antha. Aquarium (Klavins UW) open-source workflow.

Industrial foundries

Ginkgo Bioworks (Boston; SPAC 2021 — ~$15B peak market cap; foundry-as-a-service for pharma, food, agro, materials clients). Zymergen (acquired by Ginkgo 2022). Synthace (Antha software). Strateos (formerly Transcriptic; remote-execution biology). Emerald Cloud Lab (Boston; full cloud lab). Multus Biotechnology (UK; cultivated meat media). Constructive Bio (UK; Chin spin-out).

Academic foundries: Imperial College London Biofoundry, Manchester SYNBIOCHEM, Edinburgh Genome Foundry, MIT-Broad Foundry, Concordia Genome Foundry (Montreal), KAIST, Tsinghua, Macquarie ARC Centre of Excellence in Synthetic Biology.

Therapeutics and human applications

Engineered cells as drugs

• CAR-T (chimeric antigen receptor T cells). Patient T cells engineered with CAR targeting CD19 (Kymriah, Yescarta, Tecartus, Breyanzi for B-cell malignancies), BCMA (Abecma, Carvykti for multiple myeloma). Engineered ex vivo with lentivirus or retrovirus; reinfused. Allogeneic “off-the-shelf” CAR-T using TALEN or CRISPR KO of TCR + CD52 (Allogene, Cellectis UCART19) emerging. ~9 FDA-approved CAR-T as of 2026.
• TCR-engineered. T-cell receptor targeting intracellular antigens via MHC presentation. Adaptimmune ADP-A2M4 (afami-cel) for synovial sarcoma — first FDA approval (Aug 2024).
• NK cell, gamma-delta T, macrophage engineering. Earlier clinical; CAR-NK potentially safer (no GvHD).
• Engineered probiotics. Synlogic SYNB1934 (engineered E. coli Nissle for phenylketonuria; converts Phe → trans-cinnamic acid in gut; collaboration with Roche). Cobimentol SYNB1353 (homocystinuria).
• In vivo CRISPR. Editas EDIT-101 (CEP290; LCA10 — first in vivo CRISPR; subretinal injection 2020); Verve VERVE-101/102 (PCSK9 ABE for cardiovascular); Intellia NTLA-2001 (in vivo CRISPR for transthyretin amyloidosis; LNP delivery; Phase 1 data 2021).

Engineered viruses

• Oncolytic. T-VEC (Imlygic; HSV-1 with deleted ICP34.5 + IL-12 insertion; melanoma; first FDA-approved oncolytic 2015).
• Gene therapy vectors. AAV2 Luxturna (RPE65; FDA 2017), AAV9 Zolgensma (SMN1; SMA; FDA 2019; $2.1Mperdose),AAV5Hemgenix(factorIX;hemophiliaB;FDA2022;$3.5M).

Stem cell engineering

iPSC derivation (Yamanaka 2006; Nobel 2012). iPSC + CRISPR + differentiation pipelines → off-the-shelf cell therapies. Companies: Vertex/CRISPR Therapeutics (VX-880 islet cells for T1D), Sana Biotechnology, BlueRock Therapeutics, Century Therapeutics, Fate Therapeutics (iPSC-NK).

Biomanufacturing of pharma proteins

Cross-link proteomics-and-mass-spec-deep. CHO-based mAb production dominates ($200B+ annual mAb market 2024); biosimilars increasing share.

Cellular agriculture

Cultivated meat

Animal-cell-derived meat without slaughter. Bioreactor-grown muscle, fat, connective tissue from primary cells or iPSC.

• Eat Just / Good Meat. Singapore approval 2020 (chicken nuggets); US FDA approval 2023.
• Upside Foods. US FDA approval 2023; pause in 2024 on capacity scale-up.
• Mosa Meat, Aleph Farms, Believer Meats (formerly Future Meat), Memphis Meats (now Upside), Wildtype, BlueNalu.
• Cost. Mosa Meat 2013 first cultivated burger ~$330,000.By2024$10-20/lb cell-only cost; far from price parity with conventional ($3-5/lb wholesale). Growth media (basal + growth factors) ~80% of cost; recombinant growth factor production is the key cost-reduction lever.

Precision fermentation

Recombinant protein production for food.

• Perfect Day (β-lactoglobulin, α-lactalbumin) in Trichoderma; whey-protein ice creams in Brave Robot.
• Impossible Foods (leghemoglobin) in P. pastoris; meat flavor.
• The Every Company (recombinant egg white).
• Onego Bio (ovalbumin in Trichoderma).
• New Culture (mozzarella casein in Trichoderma).
• Motif FoodWorks (Ginkgo spin-off; multiple proteins).

Cultivated palm oil, cocoa, coffee

C16 Biosciences (palm-oil alternative via yeast), Voyage Foods (cocoa-free chocolate), Atomo (cell-free coffee), California Cultured (cocoa cells). Early commercial.

Regulatory landscape

FDA jurisdiction (US)

• Cell and gene therapies. OTAT (Office of Therapeutic Products, formerly OTAT). IND → Phase 1/2/3 → BLA.
• Genetically engineered animals. Center for Veterinary Medicine. AquAdvantage salmon (Aqua Bounty; growth-hormone transgenic; approved 2015 after 20-year review).
• Engineered microbes for food. GRAS (Generally Recognized As Safe) self-affirmation or formal FDA notification. Impossible’s leghemoglobin received GRAS no-objection 2018.
• Engineered crops. Coordinated USDA APHIS (plant pest), EPA (PIP — plant-incorporated protectant), FDA (food safety).

EMA (Europe)

ATMP (advanced therapy medicinal products) regulation — gene therapy, somatic cell therapy, tissue-engineered products. CAT (Committee for Advanced Therapies). GMO release (Directive 2001/18/EC); CRISPR-edited crops currently fall under GMO regulation (ECJ 2018 ruling; reform pending 2024-2026).

China NMPA

Faster CAR-T approvals than US in some cases (e.g., relmacabtagene CAR-T approved 2021); biotech investment surge 2018-2022; cooling 2023-2024.

Cartagena Protocol on Biosafety

Treaty governing transboundary GMO movement. AIA (advance informed agreement) required for living modified organisms.

Safety and dual-use concerns

Biosecurity framework

• Select Agents and Toxins (US CDC/USDA). 67 agents — Ebola, anthrax, smallpox, ricin, botulinum toxin. Registration, security, training, transfer requirements.
• iGEM safety review — student competition mandatory biosafety screening; chassis whitelist.
• HHS dual-use research of concern (DURC) policy. Identifies gain-of-function and other concerning research; requires institutional review.
• DNA synthesis screening. International Gene Synthesis Consortium (IGSC) members (Twist, IDT, Eurofins, GenScript, Charles River) screen orders against curated databases of select-agent sequences. Customer-screening (CES) and sequence-screening (SS) both required by 2024 FDA-NIH framework.

Mirror life

A 2024-2026 emerging concern: chirality-reversed organisms (mirror cells, mirror proteins, mirror DNA) would be invisible to immune systems, viruses, predators, and pathogens — but uncontrollable in the environment. Esvelt-Adamala-Endy et al. 2024 Science commentary warned against developing mirror life. Ongoing international policy discussions through ASAPbio, NASEM, WHO.

Risk of accidental release vs deliberate misuse

Most synthetic biology accidents to date have been minor (greenhouse releases of GM corn, lab-acquired infections). Deliberate misuse remains theoretical at scale but technically feasible — Esvelt’s “delay-deploy” framework recommends not publishing certain gain-of-function methodology before adequate countermeasures exist.

Modeling and ML in synthetic biology

Mechanistic models

• Kinetic models. Ordinary differential equations for enzyme kinetics, gene expression, signaling. COPASI, Tellurium, BioNetGen, JWS Online.
• Stochastic. Gillespie SSA for low-copy-number systems (TF binding, plasmid copy number). StochKit, BioNetGen-Network-free.
• Multi-scale. Combine cell-level kinetics with tissue / population dynamics.

Genome-scale models

GEMs and FBA as above. ML-augmented: DLKcat (Li et al. 2022 — deep learning enzyme k_cat); EnzymeMap, BRENDA databases. Reaction-graph models (SEED, KBase, BiGG) interlink.

Generative protein design

AlphaFold2 (Jumper-Hassabis 2021 Nature); ESMFold (Rives-Meta 2022). De novo design: RFdiffusion (Watson-Baker 2023 Nature) — diffusion models hallucinate novel binding sites and proteins; ProteinMPNN (Dauparas-Baker 2022 Science) — sequence design for fixed backbone; ESM3 (EvolutionaryScale 2024) — multimodal protein language model. AlphaProteo (DeepMind 2024) — generates protein binders with high success rates. Sequence-to-function ML enables in silico DBTL prior to lab work.

Bayesian optimization for strain engineering

Multi-armed bandit and BO algorithms guide which mutations / pathway changes to try next, given expensive measurement budget. Active in industrial labs: Ginkgo, Inscripta, Lygos.

Gene drives and ecological engineering

Gene drives propagate a genetic modification through a wild population faster than Mendelian inheritance allows — typically by self-copying onto the homologous chromosome via CRISPR.

CRISPR-based gene drives

Esvelt-Church 2014 eLife proposed CRISPR gene drives. Cas9 + sgRNA + repair template integrated as one cassette; during meiosis, Cas9 cuts the unedited homolog, repair from the drive copy makes both homologs carry the drive. Drive frequency rises from <1% to >99% in ~10 generations.

Demonstrated:

• Anopheles stephiensi malaria mosquito — Gantz-James 2015 PNAS with anti-Plasmodium genes.
• Anopheles gambiae — Kyrou-Crisanti 2018 Nat Biotechnol with doublesex disruption → population suppression in cages.
• Drosophila, mouse — proof-of-concept demonstrations.

Containment and reversibility

Daisy-chain drives (multiple linked drive elements decaying over generations), split drives (drive component on a separate transgene that does not self-propagate), and ecological barriers (host-specific endogenous regulator dependence) are countermeasures against unintended ecological spread.

Target Malaria consortium (Imperial College London, Burkina Faso, Mali, Uganda) preparing first open-field gene drive deployment — anti-Anopheles gambiae targeting malaria — projected 2026-2028 after extensive regulatory approval.

Ethical and governance challenges

WHO 2014 guidance on GM mosquitoes; Convention on Biological Diversity (Cartagena Protocol) framework. African Union policy framework 2022 conditionally supportive. NIH RAC, ABRMS, and national biosafety committees.

DNA-based information storage

The information density of DNA (~10¹⁸ bits/g theoretical) makes it the densest storage medium known. Microsoft Research + UW + Twist Biosciences synthesized ~1 GB of DNA data 2019; read back with sequencing. Open challenges: synthesis cost ($),read-cost($), random access (drop a single file out of an archive), error rates (~0.01% per nt by Illumina; <1% per oligo end-to-end with error correction).

Encoding schemes

Map binary → DNA via 1:1 base coding (00=A, 01=C, 10=G, 11=T) — simple but homopolymer-prone. Goldman-Birney 2013 base-3 ternary code; Erlich-Zielinski 2017 fountain code (DNA Fountain); rotating-window encoders avoid homopolymers and GC imbalance.

Companies

Catalog Technologies (Boston), Helixworks (UK), Iridia (USA), CD Genomics, Twist Biosciences, Illumina DataStorage Alliance. Iridia and Catalog use enzymatic synthesis to lower cost; Twist photolithographic high-density chemistry.

Future as cold storage

DNA storage targets archival data (films, scientific datasets, government records) — where access latency of hours-days is acceptable in exchange for million-year shelf life at room temperature. Tape (LTO) and optical archive currently dominate; DNA archival could reach competitive economics by 2030s if synthesis cost continues to fall.

DNA-encoded libraries (DEL)

Combinatorial chemistry on a DNA tag. Each unique DNA barcode identifies a unique small molecule attached via cleavable linker. Pool ~10⁸-10¹² compounds and pan against a target → sequence enriched barcodes → identify hits.

Companies: GSK ELT, Nuevolution (acquired by Amgen), Vipergen, X-Chem (now Arrayjet), HitGen, Wuxi LabNetwork. Hit rates: 0.001-0.1% of starting library; clinical drug originators include the inhibitors of soluble epoxide hydrolase (Boehringer-Ingelheim) and RIP1 (GSK 2018).

DEL extends synthetic biology’s tagging principle to small-molecule discovery, blurring chemistry-biology boundary.

Outlook 2026-2030

• AI-protein design. RFdiffusion, AlphaProteo, ESM3 enabling de novo protein binders to arbitrary targets. Industrial deployment imminent.
• In vivo CRISPR. PCSK9, TTR, sickle-cell editing (Casgevy approved 2023) drove the proof; expanding to neuro, lung, and ophthalmic targets.
• Engineered probiotics. Synlogic, Novome, ZBiotics entering broader clinical and consumer markets.
• Mirror life. Active policy debate; technical feasibility unresolved.
• Cell-free large-scale manufacturing. Sutro, Genovis, Liberum — commercial cell-free biopharma production scaling.
• Cultivated meat cost parity. ~2030 plausible if growth-factor and media cost decreases continue.

Hands-on tools the field uses

Cloning and DNA manipulation

• Twist (gene synthesis), IDT (oligos, gBlocks), Eurofins (sequencing, oligos), Plasmidsaurus (Nanopore plasmid sequencing).
• Promega Wizard (DNA prep), Zymo Research (DNA cleanup), NEB (enzymes — Q5, Phusion, Gibson Assembly Master Mix, NEBuilder HiFi, restriction enzymes).
• Benchling, SnapGene, Geneious, ApE — sequence design.
• Addgene (~100,000 plasmids in repository; standard cloning resource).

Strain construction

• ATCC, DSMZ, JCM, NBRP, Coli Genetic Stock Center, Bacillus Genetic Stock Center — strain repositories.
• BEI Resources — pathogen and infectious agent collections.
• Yeast Genetic Stock Center — S. cerevisiae deletion collections, MoBY library.
• Cellosaurus, ATCC, ECACC, JCRB — mammalian cell line repositories.

Screening

• Tecan Spark, BMG ClarioStar, BioTek Synergy — plate readers.
• FACS — BD FACSAria, Sony MA900, Beckman MoFlo, Bio-Rad S3e.
• LC-MS — Agilent, Sciex, Thermo, Waters, Shimadzu.
• Imaging — Zeiss, Nikon, Leica, Olympus, automated high-content imaging (PerkinElmer Opera Phenix, Molecular Devices ImageXpress).

Sequencing

• Illumina (NextSeq, NovaSeq) for short-read.
• Oxford Nanopore (MinION, GridION, PromethION) for long-read and plasmid validation.
• PacBio (Sequel II, Revio) for HiFi long-reads.
• BGI/MGI DNBSEQ (G50, T7, T20) — short-read alternative.
• Element AVITI, Singular Genomics G4 — emerging short-read challengers.

Compute

• AWS HealthOmics, GCP Life Sciences, Azure Genomics, Terra (Broad), Galaxy, DNAnexus — cloud genomics platforms.
• Snakemake, Nextflow, WDL — workflow languages.
• Anaconda + Bioconda — bioinformatics package management.
• nf-core — curated Nextflow pipelines.

Plant synthetic biology

Distinct from microbial syn bio in several ways: longer generation time (months), strict regulatory landscape (USDA APHIS plant pest, EPA pesticide, FDA food), large complex genomes.

Plant transformation

• Agrobacterium tumefaciens. Natural plant pathogen with T-DNA transfer; engineered Ti plasmid carries any cargo onto host chromosomes. Standard for dicots (Arabidopsis, tobacco, soybean, tomato, potato) and increasingly monocots (rice, maize).
• Biolistic / particle bombardment. Gold/tungsten particles coated with DNA; gas-gun delivery. Used for chloroplast transformation (no Agrobacterium pathway).
• Floral dip. Arabidopsis-specific; immerse inflorescence in Agrobacterium suspension; transformants appear in F1 seed.
• Protoplast transfection / electroporation. Cell-wall-removed plant cells; PEG-mediated DNA uptake. Required for some species; regeneration challenging.

CRISPR in plants

PEG-Cas9 RNP delivery; biolistic; or Agrobacterium-delivered Cas9 + sgRNA expression. ZmCas9 (maize-optimized), AtCas9. SpCas9, Cas12a (Cpf1) particularly popular for plants (T-rich PAM, simpler sgRNA).

Edits in commercial pipeline: drought-tolerant maize (Corteva, Bayer), browning-resistant mushroom (Yang Penn State; first non-regulated CRISPR product 2016 USDA exemption), high-oleic soybean (Calyxt/Cellectis), GMO-free hornless cattle (Recombinetics; failed regulatory but route demonstrated).

Genome-edited crops in market

• Sicilian Rouge tomato (Sanatech Seed). GABA-enriched via CRISPR; Japan 2021.
• Pioneer waxy corn. High-amylopectin starch via CRISPR-KO of granule-bound starch synthase; US 2016 USDA non-regulated; commercial 2021.
• Calyxt Calyno high-oleic soybean. TALEN-edited; commercial 2019.
• Tropic Biosciences non-browning banana. CRISPR; field trials.
• Pairwise leafy greens. CRISPR-edited mustard greens with reduced pungency; consumer launch 2023.

Plant cell factories

Tobacco BY-2, Nicotiana benthamiana for transient expression (vaccines, antibodies). Medicago (Mitsubishi Tanabe, now closed) made VLP COVID vaccine in N. benthamiana. Plant-cell suspension culture for paclitaxel (Phyton Biotech), recombinant glucocerebrosidase (Protalix Carrizyme, FDA-approved Gaucher disease enzyme).

Mammalian cell engineering

Stable cell line development

For protein therapeutics:

1. Transfect parental host (CHO, HEK293) with expression cassette (typically integrated under selection — DHFR/MTX, GS/MSX, puromycin, blasticidin).
2. Single-cell clone (limiting dilution, FACS, Solentim/Cytena imaging clonality verifier).
3. Screen 100s-1000s of clones for titer, stability over 30-50 generations, product quality.
4. Master cell bank (MCB) and working cell bank (WCB) creation with stability documentation.
5. CMC characterization for regulatory filing.

Timeline: 6-12 months from transfection to MCB.

Lonza GS Xceed, Cytiva Cell Pool

Pre-engineered platform cell lines with high-productivity locus, GS or DHFR knockout, etc. Accelerate cell-line development. Lonza GS-CHO is the most-licensed (>100 commercial products).

Targeted integration

Sage Bionetworks, ATUM Bio, Selexis SUREtarget — site-specific integration via FlpO, Bxb1, PhiC31 recombinases or CRISPR HDR into pre-validated genomic loci. Reduces clone-to-clone variability and accelerates development.

Glycoengineering

• CHO Lec1, Lec8 mutants (mannose-rich, galactosylation-deficient — for SOTA work).
• GnTI knockout CHO (Patel-Stanley) — Man5 glycan; suitable for specific therapeutics.
• FUT8 knockout (afucosyl) — ADCC-enhanced antibodies (e.g., Mogamulizumab, Kyowa Kirin Potelligent).
• Glycoengineered Pichia — GlycoFi (acquired by Merck) humanized N-glycan in yeast.

iPSC manufacturing

• Tissue source: blood, skin biopsy, urine.
• Reprogramming: Sendai virus, mRNA, episomal vectors (Yamanaka factors OSKM).
• QC: pluripotency markers (OCT4, NANOG, SSEA-4, TRA-1-60), karyotype, mycoplasma, sterility, identity.
• Differentiation: protocols for hepatocyte, cardiomyocyte, neuron, beta-cell, NK, T cell, RPE, kidney organoid.
• GMP-grade manufacturing: NIH iPSC Center, Cellular Engineering Technologies, Lonza Personalized Medicine, FUJIFILM Cellular Dynamics International, ATCC.

Open hardware and DIY biology

DIYbio movement

Community labs: Genspace (NYC), BioCurious (Sunnyvale), Counter Culture Labs (Oakland), London Hackspace Bio, Open Wetware. Low-cost equipment: OpenPCR (~$600thermocycler),OpenTrons($5k liquid handler), Coral Bench centrifuge.

Open-source DNA tools

Addgene plasmid repository (free deposits, ~$80/plasmid shipping); BioBricks public domain parts; Open Plant Initiative. iGEM Foundation runs annual undergraduate competition; ~400 teams worldwide; standardized parts catalog.

Risk vs democratization

Most DIYbio is education and visualization (GFP bacteria, glow-in-dark plants). Genuine biosafety risks are low at amateur skill level — but FBI BSAT (Bioterrorism Risk Assessment Group) monitors the community for select-agent acquisition attempts.

Frontier examples — applications under development

Cell-free protein synthesis for on-demand pharma

Sutro Biopharma Xpress CF platform produces difficult bispecifics and ADCs cell-free at multi-kg scale; CF-AB drug conjugate STRO-002 entered Phase 2 ovarian cancer 2023.

Engineered bacteriophage therapy

Locus Biosciences (CRISPR-Cas3 phages for E. coli UTI), Adaptive Phage Therapeutics, Phagomed, Eligo Bioscience (microbiome-editing CRISPR phages). Active clinical trials; regulatory pathway forming.

Heat-tolerant probiotic enzymes for nutrition

DSM ProAct enhances feed-protein digestion in poultry; reduces feed nitrogen excretion.

Carbon capture by engineered cyanobacteria

LanzaTech (Skokie IL) uses Clostridium autoethanogenum to ferment CO/CO2 to ethanol; commercial plant Ghent Belgium 2022; multiple steel-mill off-gas plants 2024-2026.

Algae fuel

Solazyme/TerraVia (closed), Sapphire Energy (closed), Synthetic Genomics (closed) — algae biofuel commercialization has been hard. Niche success in algae omega-3 fatty acids (DSM, Cellana) and astaxanthin (Algatech, AlgaeCytes). Photosynthetic biofuel economics unfavorable at current oil prices.

Notable failures and lessons

• Solazyme. Heterotrophic algae lipid; ran up against commodity oil prices.
• Genomatica BDO. Bio-based 1,4-BDO — commercial at modest scale (Cargill 30 kt/yr); cost-competitive with petrochem at high oil price.
• Amyris. Multiple molecules; financial restructuring 2023 despite scientific successes (artemisinic acid, farnesene, squalane). Lesson: biology can hit titer/rate/yield targets but consumer-product marketing/distribution is hard.
• Theranos. Not strictly syn bio but illustrative — failed attempts to miniaturize multiplex blood diagnostics; held back legitimate point-of-care development by years.
• Zymergen. ML-driven foundry promised much but lacked product-market fit beyond electronics (hyaline polyimide); acquired by Ginkgo at fraction of IPO valuation.

The common lesson: scientific feasibility is necessary but not sufficient. Cost-of-goods, regulatory path, customer acquisition, manufacturing scale-up, and capital efficiency dominate commercial outcomes once the biology works.

Industry consortia and standards

Standards bodies

• SBOL (Synthetic Biology Open Language). XML-based standard for representing genetic constructs. SBOL Visual icons for parts. Implemented in Benchling, Geneious, j5/Teselagen.
• MIDS (Minimum Information about a Genetic Sequence). Disclosure standard for published genetic designs.
• MIBBI (Minimum Information about a Biological or Biomedical Investigation). Family of reporting standards.
• iGEM Registry of Standard Biological Parts. 20,000+ characterized BioBrick parts.
• OpenMTA (Open Material Transfer Agreement). Liberal terms for sharing biological materials in academic-industry collaborations.

Industry consortia

• BioFAB (US, defunct). Early biofoundry; folded into Berkeley Lab.
• EBRC (Engineering Biology Research Consortium; US). Industry-academia roadmap; policy advocacy.
• EFB (European Federation of Biotechnology). Coordinates EU efforts.
• Singapore SynBio Consortium. A*STAR + universities + companies.
• iGEM Foundation. Annual undergraduate competition; >400 teams; trained ~50,000+ students 2003-2024.

Notable courses and educational resources

• MIT 20.020 Introduction to Biological Engineering Design (Endy, Knight).
• HHMI BioInteractive resources.
• CSHL Synthetic Biology summer course.
• iGEM training modules.
• Open courseware — Stanford BIOE80, Berkeley BIOENG, ETH Synbio, Imperial College Synbio MSc.
• Bionet community for biology hackers and DIYbio enthusiasts.

Closing perspective on the field’s maturity

Synthetic biology in 2026 is no longer a single discipline. It has bifurcated into:

1. Industrial fermentation engineering — direct continuation of metabolic engineering; biofuel, bio-chemical, food, materials.
2. Therapeutic engineering — CAR-T, gene therapy, in vivo CRISPR, engineered probiotics.
3. Generative protein design — AlphaFold-era de novo proteins, often disconnected from cell-based engineering.
4. Computational protein/genome engineering and ML — drives 1-3 with active-learning DBTL.
5. DIYbio + iGEM education — pipeline producing trained workforce.
6. Foundational science — minimal cells, xenobiology, mirror life, biosafety policy.

Each branch has its own funding, journals, conferences, and regulatory frameworks. The unifying thread is the engineering-discipline ethos: design first principles, build standardized parts, test rigorously, learn from data, iterate. That ethos increasingly defines how all biological R&D is done, not just the projects labeled “syn bio.”

Selected leading research groups (2024-2026)

• George Church (Harvard). Recoded organisms, multiplex genome editing, gene drives.
• Jennifer Doudna (Berkeley). CRISPR fundamentals and clinical translation.
• Feng Zhang (Broad). Cas9, Cas12, Cas13 discovery and engineering.
• David Liu (Broad). Base editors, prime editors.
• James Collins (MIT). Genetic circuits, synthetic biology, AI for biology.
• Christopher Voigt (MIT). Cello compiler, large-scale circuit engineering.
• Pamela Silver (Harvard). Carbon fixation, bionic leaf.
• Drew Endy (Stanford). Open Plant, BioBricks, biosafety.
• Tim Lu (MIT). Engineered probiotics, CRISPR phage.
• Jay Keasling (Berkeley). Metabolic engineering, artemisinin, opioids.
• Sang Yup Lee (KAIST). Industrial metabolic engineering.
• Adam Arkin (Berkeley). Microbial communities, KBase.
• Pamela Yeh, Karmella Haynes — gene-regulatory circuit design.
• Caroline Ajo-Franklin — bioelectronics; electrochemically active bacteria.
• Wendell Lim (UCSF). Cell engineering, synNotch.
• Frances Arnold (Caltech). Directed evolution (Nobel 2018).
• John Glass, Clyde Hutchison (JCVI). Minimal cells, JCVI-syn3.0.
• Yusuke Niimi (Keio). mRNA platforms.
• David Baker (UW). RFdiffusion, protein design (Nobel 2024).
• Demis Hassabis, John Jumper (DeepMind). AlphaFold (Nobel 2024).

Frontier challenges

Beyond E. coli and yeast chassis

Most heterologous expression still uses E. coli (proteins, small molecules) or yeast (polyketides, terpenoids, complex chiral chemistry). Niches where chassis matters:

• Streptomyces. Native antibiotic/polyketide chassis; engineered for novel NRPS/PKS clusters.
• Bacillus subtilis. Secretory; food-grade.
• Yarrowia lipolytica. Lipid biosynthesis.
• Cyanobacteria. Photoautotrophic; CO₂ fixation; engineered for biofuels and chemicals.
• Mammalian cells (CHO, HEK293).
• Insect cells (Sf9).
• Mycobacterium smegmatis. Drug-target studies.
• Bacteroides thetaiotaomicron, Lactobacillus. Gut-bacterium chassis for live biotherapeutics.

Chassis choice depends on glycosylation needs (mammalian for human-like glycans), secretion (B. subtilis, Pichia), promoter availability, regulatory acceptance, and metabolite supply.

Multi-chassis communities

Engineered consortia split labor across species — one organism captures CO₂, another converts to intermediate, third makes product. Stable co-culture is hard; division-of-labor stability requires engineered cross-feeding or quorum-coupled growth control.

Mammalian metabolic engineering

Less developed than microbial; iPSC/CHO production at industrial scale is fundamentally protein-focused. Pulling small-molecule production into mammalian cells (e.g., NSAIDs, recombinant cannabinoids) early-stage.

Synthetic genomes and chromosomes

Sc2.0 (synthetic yeast); minimized JCVI-syn3.0 mycoplasma; designer-recoded E. coli Syn61.Δ3. Synthetic mammalian chromosomes (Pl@nT consortium; long-term goal). Cost of synthesis still dominates; new enzymatic platforms (Ansa, Camena, DNA Script) trying to drive it down.

---

Further reading

• Khalil, A.S., Collins, J.J. — “Synthetic biology: applications come of age” Nat Rev Genet 2010, 11:367 — foundational review.
• Endy, D. — “Foundations for engineering biology” Nature 2005, 438:449 — vision-statement for the engineering discipline.
• Stephanopoulos, G., Aristidou, A.A., Nielsen, J. — Metabolic Engineering: Principles and Methodologies, Academic Press 1998 — the field-founding textbook.
• Nielsen, J., Keasling, J.D. — “Engineering cellular metabolism” Cell 2016, 164:1185 — modern industrial-metabolic-engineering perspective.
• Cameron, D.E., Bashor, C.J., Collins, J.J. — “A brief history of synthetic biology” Nat Rev Microbiol 2014, 12:381.
• Brophy, J.A.N., Voigt, C.A. — “Principles of genetic circuit design” Nat Methods 2014, 11:508.
• Hutchison, C.A. III et al. — “Design and synthesis of a minimal bacterial genome” Science 2016, 351:aad6253 — JCVI-syn3.0 paper.