HTGAA Spring 2026
Week 2 · February 10, 2026

DNA Read, Write & Edit

Gel electrophoresis, restriction digestion, Benchling DNA design, Twist synthesis, and the read–write–edit toolkit applied to Microcin J25 and Füzi Poiesis.

In Silico No Lab Access Committed Listener
Part 0 · Required

Basics of Gel Electrophoresis

Gel electrophoresis separates DNA fragments by size through an agarose matrix under an electric field. Negatively charged DNA migrates toward the positive electrode; smaller fragments travel faster and further than larger ones. The result is a ladder-like pattern of bands that can be compared against a molecular weight marker to determine fragment sizes.

The key parameters governing separation are: agarose concentration (higher % = better resolution of small fragments), voltage (higher voltage = faster migration but reduced resolution), buffer composition (TAE or TBE), and staining method (ethidium bromide or SYBR Safe for visualization under UV). In restriction enzyme mapping, the resulting band pattern is diagnostic — specific enzymes cut at defined recognition sequences, producing predictable fragment sizes that confirm the identity and orientation of a DNA construct.

Füzi Poiesis Application

Gel electrophoresis will be used in Aim 2 to confirm the integrity of Gibson Assembly products for pFP-A, pFP-B, and pFP-C before transformation into E. coli K-12 MG1655, and to verify chromosomal deletion cassettes for ΔhisD, ΔtrpB, and ΔleuB by colony PCR.

Part 1 · Required

Benchling & In-Silico Gel Art

Lambda DNA (48,502 bp, NEB N3011) was imported into Benchling and subjected to virtual restriction enzyme digestion with seven enzymes: EcoRI, HindIII, BamHI, KpnI, EcoRV, SacI, and SalI. The resulting fragment pattern was used to design a gel art image in the style of Paul Vanouse's Latent Figure Protocol.

Digestion Results Summary
Enzyme Recognition Sites Fragment Range Pattern Type
EcoRI 5 cuts 4.4 – 23.8 kb 5 distinct bands
HindIII 6 cuts 0.5 – 25.3 kb 6 distinct bands
BamHI 5 cuts 4.1 – 24.6 kb 5 distinct bands
KpnI 2 cuts 18.5 – 29.9 kb 2 bands (unequal)
EcoRV 21 cuts 35 bp – 4.6 kb Dense cluster
SacI 2 cuts 18.0 – 30.0 kb 2 bands (unequal)
SalI 2 cuts 17.0 – 31.0 kb 2 bands (unequal)

Key observation: EcoRV dominates the pattern with 21 recognition sites distributed across the 48.5 kb genome, producing the dense cluster of small fragments (35–1921 bp) visible in the lower band region. This density creates the visual complexity required for the gel art design in the Vanouse-style protocol.

Part 2 · Optional for Global Students

Wet Lab — Not Applicable

Global Student Note

As a committed listener at the SynBio USFQ Node (Universidad de La Frontera, Temuco, Chile), I did not have access to a wet lab for this assignment. The gel art experiment was completed entirely in silico using Benchling's virtual digest and gel simulation tools. All physical lab work is deferred to Aim 2 of Füzi Poiesis, which will be conducted at UFRO's BIOREN facility in collaboration with Dr. Ricardo Veloso Bahamonde and Dra. Ana Mutis.

Part 3 · Required

DNA Design Challenge — Microcin J25

The protein chosen for this assignment is Microcin J25 (MccJ25) — the antimicrobial peptide at the core of Strain A in Füzi Poiesis. This is not an arbitrary choice: MccJ25 is the anti-coliform effector that Strain A produces to address the documented fecal contamination of Lake Budi's littoral zone (54,000 NMP/100mL of fecal coliforms detected by SEREMI de Salud in October 2024, exceeding regulatory limits by a factor of 54).

3.1 · Chosen Protein
Microcin J25 — 21-amino-acid lasso peptide

MccJ25 is encoded by the mcjA gene (structural peptide precursor) and processed by mcjB (macrolactamization) and mcjC (further modification) into a lasso topology — a threaded structure where the C-terminal tail is locked through the N-terminal ring.

The mature peptide acts by inhibiting RNA polymerase in susceptible Gram-negative bacteria, particularly Enterobacteriaceae, with nanomolar potency. Its narrow-spectrum activity makes it ideal for selective anti-coliform function without broad disruption of the lake microbiome.

3.2 · Reverse Translation
Protein → DNA

The MccJ25 precursor peptide (McjA, 69 amino acids including signal sequence) was reverse-translated using NCBI's back-translation tool, cross-referenced against the original mcjABCD operon from E. coli AY1 (Solbiati et al. 1996; GenBank accession U40667).

For Füzi Poiesis, the complete four-gene operon (mcjA, mcjB, mcjC, mcjD) is required — not just the structural gene — because mcjB and mcjC are responsible for the lasso post-translational modification, and mcjD encodes the self-immunity/export protein.

The full mcjABCD operon spans approximately 3.8 kb in the native context. For the Aim 1 cassette in Benchling, the operon was retrieved from GenBank and codon-optimized for E. coli K-12 MG1655.

3.3 · Codon Optimization
Why codon optimization is necessary

The genetic code is degenerate — most amino acids are encoded by multiple synonymous codons. However, different organisms use these synonymous codons with different frequencies, matching their tRNA pool abundances. If a gene from one organism is expressed in another, rare codons in the host create translational bottlenecks: ribosomes pause at rare codons, reducing expression efficiency.

For Füzi Poiesis, the mcjABCD operon originates from a clinical isolate; codon optimization for E. coli K-12 MG1655 using the IDT Codon Optimization Tool ensures that each codon in the four-gene operon matches the K-12 tRNA pool. Codons with frequency below 10% in K-12 were replaced with synonymous alternatives. GC content was adjusted to match the K-12 genome average of 51%.

Example: GAA→GAG (both Glu), CGA→CGC (both Arg, but CGC is 15% more frequent in K-12), TCG→AGC (both Ser, AGC is preferred in K-12).

3.4 · From Sequence to Protein
Transcription, Translation & Post-Translational Modification

The codon-optimized mcjABCD operon, driven by a constitutive sigma-70 promoter (BBa_J23119), is transcribed by E. coli's endogenous RNA polymerase into a polycistronic mRNA. Each gene's RBS (designed in RBS Calculator v2.1) independently recruits ribosomes for translation of each protein.

McjA is translated as a 69-amino-acid precursor peptide. McjB and McjC perform the critical post-translational lasso modification: McjB catalyzes macrolactamization between the N-terminal amine (Gly1) and a glutamate residue (Glu8), forming the ring; McjC threads the C-terminal tail through the ring and locks it in the lasso topology. Critical point: Linear MccJ25 is inactive. The lasso topology is essential for activity.

McjD exports the mature peptide and confers self-immunity to Strain A. The cell-dependent (in vivo) expression system is used in Aim 1 because cell-free systems lack the post-translational modification machinery required for lasso peptide formation.

Central Dogma: DNA → RNA → Protein (mcjA example)
DNA (genomic): 5'–ATG GAG ACC CGC TTC CCT CAG CAG AGC CAG CAG ACC CCG GCG AGC ACC...–3' ↓ Transcription (T→U) mRNA (polycistronic): 5'–AUG GAG ACC CGC UUC CCU CAG CAG AGC CAG CAG ACC CCG GCG AGC ACC...–3' ↓ Translation (codon→amino acid) Protein (McjA precursor, 69 aa): M–E–T–R–F–P–Q–Q–S–Q–Q–T–P–A–S–T–...

Note: Each 3-nucleotide codon codes for 1 amino acid. The ribosome reads the mRNA 5'→3', translating each codon sequentially. Stop codons (UAA, UAG, UGA) terminate translation. Post-translational modifications occur after the peptide chain is synthesized.

3.5 · GFP as Parallel Exercise
Practice workflow — reverse translation → codon opt → Benchling → Twist

In parallel with the MccJ25 design for Füzi Poiesis, this assignment used GFPuv / GFPmut3b as a model protein to practice the full reverse translation → codon optimization → Benchling → Twist workflow. GFP provides a visually verifiable output (fluorescence under UV) that simplifies validation of expression cassette function. The GFP cassette was assembled in Benchling and submitted as a practice Twist order (see Part 4).

Part 4 · Required

Twist DNA Synthesis Order

A practice Twist Biosciences synthesis order was prepared using the GFP expression cassette assembled in Benchling. The insert was built as a linear expression cassette including: Promoter (BBa_J23106), RBS (BBa_B0034 with spacers), Start codon, Coding sequence (GFPmut3b, codon-optimized for E. coli), 7× His-tag, Stop codon, and Terminator (BBa_B0015).

pTwist Amp High Copy + GFP Insert
Construct Summary

Vector: pTwist Amp High Copy — ampicillin resistance, ColE1 high-copy origin

Insert: 924 bp GFP expression cassette — promoter, RBS, GFPmut3b, His-tag, terminator

Total construct: 3,145 bp

Complexity: Standard

Estimated cost: $118.16 USD

Order type: Clonal gene (circular DNA) — directly transformable into E. coli without additional assembly steps.

GFP Expression Cassette Structure

Promoter (BBa_J23106): 5'–TTTACGGCTAGCTCAGTCCTAGGTATAGTGCTAGC–3'

RBS (BBa_B0034 + spacers): 5'–CATTAAAGAGGAGAAAGGTACC–3'

Start Codon: 5'–ATG–3'

GFPmut3b (codon-optimized, 724 bp / 241 aa): 5'–AGCAAAGGAGAAGAACTTTTCACTGGAGTTGTCCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCAC...–3'

7× His-tag: 5'–CATCACCATCACCATCATCAC–3'

Stop Codon: 5'–TAA–3'

Terminator (BBa_B0015): 5'–CCAGGCATCAAATAAAACGAAAGGCTCAGTCGAAAGACTGGGCCTTTCGTTTTATCTGTTGTTTGTCGGTGAA...–3'

Füzi Poiesis Connection

For the actual Füzi Poiesis synthesis in Aim 2, Twist would be used to synthesize the three deletion cassettes (ΔhisD, ΔtrpB, ΔleuB — ~1.5 kb each with homology arms) and the pFP-C AND-gate expression cassette (4,238 bp). The GFP practice order here demonstrates the complete workflow: Benchling annotation → FASTA export → Twist upload → vector selection → construct download → re-import to Benchling for verification.

Part 5 · Required

DNA Read / Write / Edit

5.1 · DNA Read
What DNA would I sequence and why?

Target: The benthic metagenome of Lake Budi's anoxic sediment layer (below 4.5 m depth, where dissolved oxygen ≤ 0 mg/L and negative redox potential is documented). This metagenome does not currently exist in the published literature — a systematic bibliographic analysis of 77 documents on the Budi basin (LME-UChile, 2010) identified microbiology as the least-represented research topic.

Why this DNA matters for Füzi Poiesis: The horizontal gene transfer risk from the synthetic consortium to native lake organisms depends entirely on what genes exist in the native metagenome. If native Lake Budi bacteria carry intact hisD, trpB, or leuB homologs, HGT could restore auxotrophic independence in an escaped strain — collapsing the primary biocontainment mechanism. Without metagenome data, this risk cannot be quantified.

Technology chosen: Oxford Nanopore Technology (ONT) long-read sequencing (third-generation), specifically the MinION platform. Reason: long reads (N50 > 20 kb) resolve repetitive regions and enable de novo metagenome assembly without a reference genome. The MinION's portability is critical — it can be deployed at the field sampling site in Puerto Domínguez, eliminating the cold-chain transport of environmental DNA samples to Temuco or Santiago. Real-time base calling via Guppy allows same-day quality assessment in the field.

Input preparation: High-molecular-weight DNA extraction from sediment cores using a bead-beating + CTAB protocol. Size selection via BluePippin to enrich fragments > 10 kb. Library preparation using ONT's Ligation Sequencing Kit (SQK-LSK114). No PCR amplification — metagenome-wide sequencing without amplification bias.

5.2 · DNA Write
What DNA would I synthesize and why?

Target constructs for Füzi Poiesis Aim 2:

ΔhisD deletion cassette (~1.5 kb) — genomic deletion of histidine biosynthesis gene in Strain A. Synthesized as gene fragments with 500 bp homology arms flanking the hisD locus in E. coli K-12 MG1655, compatible with lambda Red recombineering.
ΔtrpB deletion cassette (~1.5 kb) — tryptophan auxotrophy in Strain B, same architecture.
ΔleuB deletion cassette (~1.5 kb) — leucine auxotrophy in Strain C.
pFP-C complete plasmid (4,238 bp) — the AND-gate PhoA expression construct with P_lux, P_phoB, RBS, codon-optimized phoA, terminators, ts-ori, and selection marker, to be ordered as a clonal gene in a Twist vector.

Technology: Twist Biosciences phosphoramidite-based gene synthesis. Gene fragments for deletion cassettes (linear, with exonuclease protection) to enable direct use in lambda Red recombineering. Clonal gene for pFP-C (circular) for direct transformation. Synthesis verified by Sanger sequencing of 3–5 colonies post-transformation before proceeding to auxotroph selection.

Limitations of Twist Synthesis

Accuracy: Twist achieves ~99.98% per-base accuracy (theoretical error rate of ~1 per 10,000 bp). For the ~1.5 kb deletion cassettes, this translates to high fidelity, though Sanger verification post-synthesis is standard practice.

Scalability: Phosphoramidite synthesis works efficiently up to ~10 kb constructs. Beyond ~15 kb, synthesis times increase significantly, and multi-fragment assembly becomes more economical. For pFP-C (4.2 kb) and deletion cassettes (1.5 kb), no assembly is required post-synthesis.

Turnaround: Standard Twist orders ship in 2–3 weeks. Expedited options (1 week) are available at additional cost. For Füzi Poiesis Aim 2 timeline, standard turnaround is acceptable.

5.3 · DNA Edit
What DNA would I edit and why?

Target edits for Füzi Poiesis Aim 2: Chromosomal deletion of hisD, trpB, and leuB in E. coli K-12 MG1655 (three separate strains) to create the auxotrophic ring. Chromosomal integration is required — plasmid-based auxotrophy can be lost through plasmid curing, which would break the biocontainment mechanism.

Technology chosen: Lambda Red Recombineering (not CRISPR). Reason: lambda Red recombineering is the gold-standard method for precise chromosomal deletions in E. coli K-12 and does not require a PAM site. The Datsenko-Wanner protocol (2000) enables single-step replacement of any chromosomal locus with a selectable marker flanked by FLP recombinase sites, which can subsequently be excised to leave a clean scarless deletion.

Essential steps:

  1. Transform target strain with pKD46 (lambda Red helper plasmid, temperature-sensitive);
  2. Induce lambda Red genes (exo, bet, gam) with arabinose;
  3. Electroporate linear deletion cassette (synthesized by Twist) with 500 bp homology arms;
  4. Plate on selective media; screen by colony PCR with flanking primers;
  5. Confirm deletion by Sanger sequencing;
  6. Cure pKD46 at 37°C (non-permissive temperature for ts-ori).

Limitations: Recombineering efficiency in E. coli K-12 is ~10⁻⁶ to 10⁻⁵ per cell per electroporation — requiring screening of 50–200 colonies per deletion. Three deletions in three separate strains requires three independent recombineering cycles. In native Lake Budi halotolerant strains (Aim 2, Level 2), lambda Red may not function without adaptation; alternative methods (CRISPR-Cas12a or I-SceI counterselection) are documented alternatives for non-model organisms.

The Week

February nights, and learning from giants

Week 1 finished late one night before heading to Chiloé — 18 hours south. A week there, then back. Week 2 done the day I returned. February in Temuco is brutal: 35°C and unlivable during the day, so you work at night when it finally cools down. But Chiloé was different. I came back with my heart full.

That matters when you're in a course taught by George Church, David Kong, and others who are actively writing this field. There's a specific humility in that — not paralyzing, but real. It makes you pay attention. During that week, I also got in touch with the SynBio USFQ node, the Latin American hub. It's where the work is actually happening.

So I stayed up that night. Not because the work was easy — the trip was exhausting, catching up on lectures and recitations and then completing Week 2 all in one go meant no sleep. But I believe in this course enough to do it anyway. That's what it means to sit in a room with giants.

Reflection

What this week grounded

Week 2 made the read–write–edit toolkit concrete in a way the Füzi Poiesis design document had treated abstractly. Designing the GFP practice cassette in Benchling — annotating every part, simulating the restriction digest, building the Twist order — revealed a gap between computational design fluency and experimental design fluency. I can write the six-ODE model; I had not fully thought through what the lambda Red recombineering protocol looks like for three simultaneous deletions in three strains.

The metagenome sequencing question (5.1) was the most clarifying. The entire HGT risk calculation in Aim 1's biocontainment validation is conditional on a metagenome that doesn't exist. This is acknowledged in the project document as a limitation — but this week's assignment made clear that it's not just a scientific gap. It's a governance gap: without metagenome data, no FPIC process can honestly represent the ecological risk of field deployment to Lafkenche communities. Sequencing Lake Budi's benthic metagenome is therefore not just a prerequisite for Aim 2 — it's a prerequisite for meaningful community consent.