Week 9 HW: Cell-Free Systems
Week 9 — Cell-Free Systems
Constantin Convalexius · Lifefabs Node · HTGAA 2026
Lecturers: Kate Adamala, Peter Nguyen, Ally Huang
Part A — General Questions
1. Advantages of Cell-Free Protein Synthesis Compared With In Vivo Expression
Cell-free protein synthesis, often shortened to CFPS, means making protein outside living cells. Instead of transforming bacteria or mammalian cells and asking the cells to produce a protein, we use a reaction mixture that contains the useful molecular machinery from cells: ribosomes, polymerases, tRNAs, amino acids, salts, cofactors, and an energy system.
The big advantage is control. In a living cell, many variables are hidden or hard to tune because the cell is trying to survive. In a cell-free reaction, the experimenter can directly control magnesium, potassium, DNA concentration, amino acids, redox state, additives, and reaction time.
Three major advantages are:
- Direct access to the reaction. Reagent concentrations such as Mg2+, K+, NTPs, amino acids, pH, redox state, and DNA template can be tuned quickly without growth and induction cycles.
- No membrane and no cell viability constraint. Toxic proteins can be expressed because there is no living host that needs to survive.
- Speed and parallelization. Results can appear within 4-48 hours, and reactions can be miniaturized into 96-well or 384-well plates.
Two cases where CFPS is more beneficial than in vivo expression are:
- High-throughput design-build-test screening. A 384-well run can test many promoter, RBS, template, or reaction-condition variants in parallel. Doing the same experiment in cells would require transformation, colony picking, growth, induction, and measurement for every variant.
- Toxic protein expression. Antimicrobial peptides such as melittin, LL-37, or colicins may kill the host cells that express them. In CFPS, there is no host cell to kill, so toxic products are easier to produce and study.
2. Main Components of a Cell-Free Expression System
| Component | Role |
|---|---|
| Cell extract / lysate | Provides ribosomes, translation factors, aminoacyl-tRNA synthetases, chaperones, and many metabolic enzymes. This is the biological engine of the reaction. |
| RNA polymerase | Transcribes DNA into mRNA. T7 RNA polymerase is commonly used when the DNA template has a T7 promoter. |
| DNA template | Contains the genetic instruction for the protein to be produced. It can be plasmid DNA or linear DNA. |
| NTPs | Building blocks for RNA synthesis during transcription. |
| Amino acids | Building blocks for protein synthesis during translation. |
| Energy regeneration system | Regenerates ATP and GTP so transcription and translation can continue for hours instead of minutes. |
| Mg2+ | Essential cofactor for ribosomes, polymerases, and ATP-utilizing enzymes. It is often one of the most important variables to optimize. |
| K+ | Helps maintain ionic strength and supports ribosome function. |
| Buffer | Keeps the reaction pH stable, commonly around physiological pH. |
| Polyamines such as spermidine and putrescine | Help stabilize nucleic acids, tRNA, and ribosome function. |
| Optional additives | Examples include DTT for reducing conditions, PEG for molecular crowding, RNase inhibitors, chaperones, or detergents/nanodiscs for membrane proteins. |
3. Why Energy Regeneration Is Critical
Protein synthesis is energetically expensive. Each amino acid added to a growing protein chain costs high-energy phosphate bonds: ATP is used to charge amino acids onto tRNAs, and GTP is used during translation elongation and translocation. Without energy regeneration, ATP and GTP would be depleted quickly, and the reaction would stop.
A practical energy-regeneration strategy is to use a system such as phosphoenolpyruvate (PEP) plus pyruvate kinase or a more sustained system based on glucose metabolism in the lysate. For longer reactions, I would prefer a glucose or ribose-supported energy system because it can feed endogenous metabolic enzymes and maintain ATP production over many hours.
This matters for my final project because the planned Ginkgo Cloud Lab experiment depends on enough protein being produced over the cell-free reaction time for split-GFP fluorescence to become detectable.
4. Prokaryotic Versus Eukaryotic Cell-Free Systems
| System | Protein I Would Produce | Why |
|---|---|---|
| Prokaryotic CFPS, such as E. coli BL21 lysate | A soluble reporter such as sfGFP or my PARP1 catalytic-domain fusion construct | E. coli lysate is fast, inexpensive, high-yielding, and well matched to T7-driven expression. It is a good first choice for simple soluble proteins or domains that do not require eukaryotic post-translational modifications. |
| Eukaryotic CFPS, such as wheat germ, rabbit reticulocyte, or HeLa lysate | A mammalian regulatory protein such as phosphorylated p53 | Eukaryotic systems are better when the protein needs eukaryotic folding machinery, post-translational modifications, or mammalian cofactors. Bacterial CFPS may produce the sequence but not the biologically relevant form. |
The tradeoff is that bacterial systems are usually cheaper and higher-yielding, while eukaryotic systems can better represent mammalian protein biology.
5. Designing a Cell-Free Experiment for a Membrane Protein
Membrane proteins are challenging in CFPS because their hydrophobic transmembrane helices usually need a lipid-like environment. Without a membrane mimic, the hydrophobic parts can aggregate or misfold.
For a membrane-protein CFPS experiment, I would:
- Add nanodiscs made from membrane scaffold proteins and lipids such as POPC. Nanodiscs provide small soluble membrane patches.
- Test small unilamellar vesicles as an alternative lipid environment.
- Add folding helpers such as GroEL/GroES and DnaK/DnaJ if the lysate does not provide enough chaperone activity.
- Use a fluorescent fusion or activity assay to detect whether the protein is folded and functional.
- Run a condition screen varying lipid composition, Mg2+, temperature, and DNA concentration.
The readout would depend on the protein. For a transporter, I could use substrate uptake into vesicles. For a receptor, I could use ligand binding. For a fluorescently tagged membrane protein, I could compare fluorescence in soluble and pellet fractions to estimate aggregation.
6. Low Yield Troubleshooting
| Possible reason for low yield | Troubleshooting strategy |
|---|---|
| Mg2+ is not at the optimum concentration | Titrate Mg2+ across a range, for example 4-16 mM. Magnesium strongly affects ribosomes and energy metabolism, so small changes can matter. |
| DNA template is degraded | Use circular plasmid instead of linear DNA, verify the template on a gel, and consider nuclease-reduced lysate or protective DNA-end modifications for linear templates. |
| Rare codons slow translation | Codon-optimize the sequence for E. coli, supplement rare tRNAs, or use a lysate made from a strain enriched for rare tRNAs. |
| Protein misfolds or aggregates | Lower the temperature, reduce DNA concentration, add chaperones, shorten the construct, or test solubility tags. |
| mRNA is unstable | Use a strong 5’ UTR/RBS design, add RNase inhibitors, and avoid long untranslated regions or unstable RNA structures. |
Kate Adamala — Synthetic Minimal Cell Design
My Design: LactoLyse, a Lactate-Sensing TRAIL-Releasing Synthetic Cell
The synthetic minimal cell I propose is called LactoLyse. It senses high extracellular L-lactate, which is common in highly glycolytic tumor microenvironments, and releases the apoptosis-inducing ligand TRAIL.
1. Function
What would the synthetic cell do?
The input is high L-lactate, for example above 5 mM. The output is production and release of TRAIL, a protein that can trigger apoptosis in susceptible cancer cells.
Could this be done by cell-free TX/TL alone without encapsulation?
Not as cleanly. Without encapsulation, TRAIL would be produced and diffuse from the start. Encapsulation creates a boundary, so the synthetic cell can act more like a local sensor-and-release device.
Could this be done with a genetically modified natural cell?
Yes, in principle. For example, engineered immune cells or engineered bacteria could sense lactate and secrete a therapeutic protein. However, living engineered cells introduce extra risks such as immune reactions, proliferation, mutation, persistence, and harder biocontainment.
Desired outcome
In lactate-rich tumor-like conditions, the synthetic cell should activate TRAIL production and release. In normal low-lactate conditions, it should remain mostly silent.
2. Components
Membrane
The membrane would be made from POPC and cholesterol. POPC forms the lipid bilayer, and cholesterol improves membrane stability and reduces nonspecific leakage.
Encapsulated contents
The vesicle would contain bacterial CFPS, a DNA template encoding TRAIL, a lactate-responsive regulatory element, amino acids, NTPs, salts, and an energy regeneration system.
TX/TL source
I would use bacterial E. coli lysate because it is cheap, fast, and compatible with many riboswitch-style regulatory designs.
Communication with the environment
Small lactate molecules can diffuse or be transported into the vesicle. For stronger control, a lactate transporter or pore-forming system could be included. Release of the protein output could be coupled to expression of a pore-forming protein such as alpha-hemolysin.
3. Experimental Details
Possible components:
- Lipids: POPC and cholesterol.
- Gene 1: human TNFSF10, which encodes TRAIL.
- Gene 2: Staphylococcus aureus alpha-hemolysin (hla) as a pore-forming release system.
- Regulatory element: lactate-responsive RNA or transcriptional control element.
- TX/TL system: E. coli BL21 Star cell-free lysate.
To measure function, I would test vesicles in low-lactate and high-lactate media. I would measure TRAIL release by ELISA and use a fluorescent reporter version in early optimization. In a cell-culture assay, I would compare apoptosis in tumor-like cells versus non-tumor control cells using Annexin V staining or a viability assay.
Peter Nguyen — Freeze-Dried Cell-Free Systems in Materials
Field: Textiles and Wound Care
Pitch
WoundMesh is a freeze-dried cell-free wound dressing that synthesizes an antimicrobial peptide on demand when activated by wound fluid.
How It Works
A wound dressing would be embedded with freeze-dried CFPS pellets containing a T7-driven antimicrobial peptide expression cassette, cell-free lysate, amino acids, salts, and an energy system. In the package, the dressing is dry and inactive. When placed on a wet wound, wound exudate rehydrates the pellets and starts protein expression.
The antimicrobial peptide could be LL-37, a human cathelicidin peptide with broad antimicrobial activity. The protein would be produced locally at the wound site, reducing the need for systemic antibiotic exposure. A simple color reporter could be included to show that the dressing has activated.
Societal Challenge
Chronic wound infections are a major problem in diabetic foot ulcers, burns, and post-surgical wounds. Systemic antibiotics can cause side effects and contribute to antimicrobial resistance. A local, disposable, on-demand antimicrobial dressing could reduce systemic exposure while still treating the infected tissue environment.
Addressing Limitations
- Activation with water: wound fluid provides the water needed to start the cell-free reaction.
- Stability: freeze-drying can make the system shelf-stable at room temperature.
- One-time use: wound dressings are already single-use, so the one-shot nature of CFPS fits the application.
- Dose control: the amount of DNA and lysate dried into the dressing sets the maximum protein dose.
Ally Huang — Mock Genes-in-Space Proposal
Topic: Real-Time Biomarker Monitoring of Microgravity-Induced Muscle Atrophy
1. Background
Astronauts lose skeletal muscle mass in microgravity because muscles are unloaded for long periods. This is a major challenge for long-duration missions to the Moon or Mars. Molecular markers such as MuRF1 and Atrogin-1 increase early during muscle atrophy, before large visible changes occur. A simple in-flight biosensor could help astronauts monitor muscle loss and adjust exercise or nutrition countermeasures.
2. Molecular Target
MuRF1 (TRIM63) and Atrogin-1 (FBXO32) mRNA, two muscle-specific markers associated with muscle protein degradation and atrophy.
3. Connection to the Space Biology Challenge
Microgravity reduces mechanical load on muscle, which activates pathways that break down muscle proteins. MuRF1 and Atrogin-1 are E3 ubiquitin ligases that target muscle proteins for degradation. Their mRNA levels rise early during unloading, so they are useful early-warning biomarkers. A freeze-dried cell-free biosensor could detect these RNAs without requiring a full molecular biology lab in space.
4. Hypothesis / Research Goal
My hypothesis is that freeze-dried BioBits cell-free reactions containing RNA toehold-switch biosensors can detect increased MuRF1 and Atrogin-1 mRNA in muscle-derived samples. If the target RNA is present, the toehold switch opens and allows translation of a fluorescent reporter.
The goal is to create a low-mass, low-power diagnostic system for spaceflight. The system should be stable at room temperature, activated only when rehydrated, and readable with a simple fluorescence viewer. This would make it easier to monitor astronaut muscle health during long missions.
5. Experimental Plan
I would test four sample types in triplicate:
- RNA from untreated C2C12 muscle cells.
- RNA from dexamethasone-treated C2C12 cells as an atrophy positive control.
- RNA from simulated-microgravity muscle cultures.
- Buffer-only no-template control.
Each sample would be added to freeze-dried BioBits reactions containing MuRF1 and Atrogin-1 toehold switches linked to fluorescent reporters. Fluorescence would be measured at 30, 60, and 120 minutes. A successful result would show at least three-fold signal over negative controls.
Part B — Individual Final Project Aim 1
Aim 1: Build and Test a Cell-Free PARP1-HPF1 Split-GFP Biosensor
My final project Aim 1 is to build and test a cell-free split-GFP biosensor for the PARP1-HPF1 protein interaction. I designed three Twist clonal gene constructs:
- PARP1cat-WT-GFP11: PARP1 catalytic domain, wild type, His6-tagged, fused to GFP11.
- PARP1cat-E988K-GFP11: PARP1 catalytic domain with E988K mutation, His6-tagged, fused to GFP11.
- HPF1-GFP1-10: full-length HPF1, His6-tagged, fused to GFP1-10.
The direct readout is split-GFP fluorescence after co-expression in a Ginkgo Cloud Lab E. coli cell-free system. If HPF1 and PARP1 come close together through binding, GFP1-10 and GFP11 can reassemble and produce green fluorescence.
This Aim 1 is intentionally scoped as a construct design and biosensor validation project. It does not directly measure PARP1 catalytic activity or cellular reprogramming. Those would require additional future assays, such as a PARylation assay, NAD+ depletion assay, mammalian cell experiments, RNA-seq, or epigenetic clock measurements.
The strongest honest claim is: I am building a molecular tool that can report PARP1-HPF1 proximity in a cell-free reaction. If it works, it becomes a foundation for future, more complete tests of scaffolding mechanisms in reprogramming regulators.
Sources
- Lentini et al., Nature Communications 2014, 5:4012.
- Pardee et al., Cell 2016, 165:1255-1266.
- Pardee, Green, Yin et al., Cell 2014, 159:940-954.
- Adamala lab synthetic-cell publications.
- BioBits and miniPCR Genes in Space materials.
- Cabantous, Terwilliger, and Waldo, Nature Biotechnology 2005, split-GFP method.
- Suskiewicz et al., Nature 2020, HPF1-PARP1 interaction.