Week 3 HW: Cell-Free Protocols
Introduction
This homework focuses on designing cell-free protein synthesis (CFPS) reaction compositions for sfGFP, PETase, and Reteplase, with the goal of improving functional protein production while reducing cost per gram. In CFPS, transcription and translation occur outside living cells, which makes the reaction highly tunable and well suited to rapid, automated screening.
This connects closely with my current research on mycelium–AgNP biomanufacturing. Although the products are different, both systems involve the same general challenge: understanding how reaction or process conditions influence kinetics, structure, function, and overall process efficiency.
For me, HW3 is therefore not only about optimizing a CFPS recipe. It is also a useful model for thinking about how to design and automate multivariable biomanufacturing experiments, which is directly relevant to my longer-term goal of building predictive workflows for mycelium–nanoparticle systems.
The Shared Research Logic is as follows.
| HW3 CFPS | My mycelium–AgNP research |
|---|---|
| Reaction composition | Fungal/process environment |
| Mg²⁺, K⁺, energy source, redox components | Ag⁺ concentration, medium, physiological state, redox chemistry |
| Protein synthesis kinetics | Ag⁺ depletion / AgNP formation kinetics |
| Protein folding/function | Nanoparticle structure / antimicrobial function |
| Functional protein yield | Functional nanocomposite performance |
| Cost per gram protein | Process efficiency / material yield / cost |
| 384-well automated screening | Future automated microreactor / DOE workflow |
| AI-designed formulations | Predictive modelling / agent-assisted experiment design |
The assignment also uses a staged 384-well → 96-well → 2 mL screening strategy and an automated platform to compare many reaction compositions efficiently. This is especially relevant to my research direction because I want to apply similar design-build-test-learn and automation strategies to study how fungal state and process conditions govern Ag⁺ reduction, AgNP formation, and material function.
A.1 Potassium Glutamate vs. Magnesium Glutamate
Both salts help establish the ionic environment required for CFPS. The homework treats potassium and magnesium salts as variable components of the reaction composition.
| K⁺ — Ionic Environment | Mg²⁺ — More Sensitive Tuning Variable |
|---|---|
| K⁺ ↓ Bulk monovalent cation ↓ Ionic strength / translation environment ↓ Protein synthesis | Mg²⁺ ↓ Ribosome structure + tRNA/ribosome interactions + nucleotide complexes + enzyme activity ↓ Protein synthesis |
| Mainly tunes the overall ionic environment and generally has a broader useful concentration range. | Directly participates in several molecular processes, so relatively small concentration changes can have a larger effect on CFPS performance. |
Their optimal concentrations differ between formulations because other reaction components, especially nucleotides and energy-system components, interact with Mg²⁺ and change how much is functionally available.
Which is more sensitive?
Mg²⁺ is generally the more sensitive variable, while K⁺ is also important but usually acts over a broader operating range.
Connection to My Research
The same principle also appears in my mycelium–AgNP system: the amount of an ion added to a reaction is not necessarily the same as the amount that is chemically or functionally available.
| CFPS — Mg²⁺ | My Mycelium–AgNP Research — Ag⁺ |
|---|---|
| Total Mg²⁺ ↓ Interaction with nucleotides and other components ↓ Functionally available Mg²⁺ ↓ Protein synthesis | Total Ag⁺ ↓ Interaction with biomolecules and medium components ↓ Chemically available Ag⁺ ↓ Ag⁺ reduction / AgNP formation |
| The CFPS response depends on how much Mg²⁺ remains available to support translation. | AgNP formation depends on how much Ag⁺ remains available for reduction within the fungal reaction environment. |
Key connection: in both systems, performance depends not only on the total concentration added, but also on how that component interacts with the surrounding biochemical environment.
What I can learn: In my research, this means I should consider how medium composition, biomolecules, and fungal metabolism influence Ag⁺ availability, rather than treating Ag⁺ concentration as an isolated variable.
A.2 PEP + NTP vs. Ribose + NMP
| System | Main components | What they provide | How the CFPS system uses them | Simple way to think about it |
|---|---|---|---|---|
| PEP + NTP | PEP = phosphoenolpyruvate; NTPs = ATP, GTP, CTP, UTP | PEP provides an energy-regeneration substrate; NTPs are ready-to-use RNA building blocks | Requires relatively little conversion before transcription can proceed | Direct supply: give the reaction energy + activated nucleotides |
| Ribose + NMP | Ribose = sugar/carbon source; NMPs = AMP, GMP, CMP, UMP | Provides cheaper metabolic/nucleotide precursors | Relies more on lysate metabolism to regenerate energy and convert NMPs into usable NTPs | Metabolic supply: give cheaper precursors and let the lysate rebuild what it needs |
| PEP + NTP | Ribose + NMP |
|---|---|
| PEP + NTP ↓ Direct energy + activated nucleotides ↓ Ready-to-use NTPs ↓ Transcription + translation | Ribose + NMP ↓ More metabolic conversion ↓ NMP → NDP → NTP ↓ Transcription + translation |
| More direct: the reaction receives a high-energy substrate and nucleotides already in the triphosphate form needed for RNA synthesis. | More metabolism-dependent: the lysate must regenerate energy and convert cheaper NMP precursors into usable NTPs. |
Key difference: PEP + NTP provides more of the required chemistry directly, while Ribose + NMP relies more on residual lysate metabolism and can reduce reagent cost.
Connection Back to A.1
This also explains why the optimal Mg²⁺ concentration can change between CFPS formulations.
| PEP + NTP | Ribose + NMP |
|---|---|
| More NTPs are supplied directly | More NTPs are generated from NMPs inside the lysate |
| NTPs bind Mg²⁺ | Mg²⁺ demand changes as nucleotides are regenerated |
| Free Mg²⁺ can decrease | Free Mg²⁺ can follow a different balance |
| Mg-Glu may need a different optimum | Mg-Glu may need a different optimum |
Connection to A.1: This reinforces why Mg²⁺ is a sensitive tuning variable: its functionally available concentration depends on interactions with nucleotides and other reaction components, so changing the energy/nucleotide system can also shift the optimal Mg²⁺ level.
Connection to My Research
A similar interaction occurs in my mycelium–AgNP system.
| CFPS | My Mycelium–AgNP Research |
|---|---|
| Energy / nucleotide system changes | Medium composition / fungal physiological state changes |
| Changes Mg²⁺ binding and availability | Changes Ag⁺ binding, reduction, and chemical availability |
| Functionally available Mg²⁺ strongly influences translation performance | Chemically available Ag⁺ influences Ag⁺ reduction kinetics and AgNP formation |
| Optimal Mg-Glu depends on the whole formulation | Optimal Ag⁺ response depends on the whole biochemical environment |
| CFPS | My Mycelium–AgNP Research |
|---|---|
| Energy / nucleotide system ↓ Mg²⁺ binding changes ↓ Functionally available Mg²⁺ ↓ Protein synthesis | Medium + fungal metabolism ↓ Ag⁺ interactions / reduction chemistry ↓ Chemically available Ag⁺ ↓ Ag⁺ depletion + AgNP formation |
| Changing the formulation changes how much Mg²⁺ remains available to support translation. | Changing the biochemical environment may change how much Ag⁺ remains available for reduction and nanoparticle formation. |
What I can learn: in both systems, the amount initially added is not necessarily the amount functionally available. Therefore, I should consider interactions between process variables and ion availability over time, rather than optimizing Ag⁺ concentration as an isolated variable.
A.3 Comparison of Two Literature CFPS Formulations
I selected PANOx-SP from Jewett & Swartz (2004) and the RFopt formulation from Olsen et al. (2026). These represent two different approaches to supporting cell-free protein synthesis.
Composition Comparison
| Component / Function | PANOx-SP — Jewett & Swartz (2004) | RFopt — Olsen et al. (2026) |
|---|---|---|
| K-glutamate | 175 mM | 362 mM |
| Mg-glutamate | 20 mM | 8 mM |
| NH₄-glutamate | 10 mM | — |
| Energy / carbon source | PEP, 33 mM | Glucose, 10 mM + Ribose, 50 mM |
| Nucleotides | ATP 1.2 mM; CTP/GTP/UTP 0.85 mM each | AMP 3 mM; CMP/GMP/UMP 2.15 mM each |
| Amino acids | 2 mM each | 5 mM amino-acid mixture* |
| Buffer | No added HEPES in the PANOx-SP additive set | HEPES, 75 mM |
| Phosphate | — | 15 mM |
| Polyamines | Spermidine 1.5 mM + Putrescine 1 mM | — |
| Other cofactors | NAD, CoA, folinic acid, tRNA | Nicotinamide, 4 mM |
| Oxalate | 2.7 mM | — |
*Reported as the amino-acid component in the RFopt formulation.
The two formulations therefore differ not only in individual concentrations, but also in their overall strategy for supplying ions, energy, nucleotides, and cofactors.
Overall Formulation Strategy
| PANOx-SP | RFopt |
|---|---|
| PEP as a high-energy substrate | Glucose + ribose as metabolic substrates |
| NTPs supplied directly | NMPs supplied as precursors |
| More externally supplied cofactors / translation components | Fewer added expensive cofactors |
| More direct biochemical supply | Greater use of residual lysate metabolism |
Comparison: PANOx-SP uses PEP and ready-to-use NTPs, together with several added cofactors. In contrast, RFopt uses glucose, ribose, and NMPs, relying more on metabolic activity in the lysate to regenerate energy and nucleotides.
Main difference: PANOx-SP represents a more direct-supply formulation, while RFopt is a more simplified, metabolism-dependent, and cost-optimized formulation.
Connection to My Research
The PANOx-SP vs. RFopt comparison is similar to how I think about my mycelium–AgNP system.
| CFPS formulation design | My mycelium–AgNP research |
|---|---|
| Directly supply many reaction components | Add Ag⁺ and defined medium/process inputs |
| Rely more on residual lysate metabolism | Rely on fungal metabolism and biomolecules for Ag⁺ reduction |
| Performance depends on the whole formulation, not one reagent | AgNP formation depends on the whole biochemical environment, not Ag⁺ concentration alone |
| Optimize for yield + cost + robustness | Optimize for Ag⁺ conversion + nanoparticle properties + antimicrobial function + process efficiency |
What I can learn: a good formulation does not necessarily need more components; it needs the right combination of inputs and biological activity. For my research, this supports studying interactions between medium composition, fungal physiological state, Ag⁺ availability, and process conditions, rather than optimizing each factor independently.
Part B: First-Round CFPS Formulations
The following were my first-round 20 µL CFPS formulations submitted to the HTBAA coopetition. Values are final concentrations in the complete reaction, including contributions from the fixed base buffer.
| Component | sfGFP | TfCut2 / PETase | Reteplase |
|---|---|---|---|
| Reaction regime | Reducing | Reducing | Oxidizing / DSB |
| K-glutamate | 300.625 mM | 300.625 mM | 300.625 mM |
| Mg-glutamate | 8.225 mM | 8.225 mM | 8.225 mM |
| HEPES-KOH, pH 7.5 | 60 mM | 75 mM | 75 mM |
| K-phosphate | 7.5 mM | 7.5 mM | 7.5 mM |
| Glucose | 1.5 g/L (~8.3 mM) | 1.5 g/L (~8.3 mM) | 1.5 g/L (~8.3 mM) |
| Ribose | 6.0 g/L (~40 mM) | 6.0 g/L (~40 mM) | 6.0 g/L (~40 mM) |
| AMP | 1 mM | 1 mM | 1 mM |
| CMP | 1 mM | 1 mM | 1 mM |
| GMP | 1 mM | 1 mM | 1 mM |
| UMP | 1 mM | 1 mM | 1 mM |
| 17-AA mix | 4 mM | 4 mM | 4 mM |
| Tyrosine | 4 mM | 4 mM | 4 mM |
| Cysteine | 4 mM | 4 mM | 1 mM |
| Nicotinamide | 4 mM | 4 mM | 4 mM |
| GSSG | — | — | ~4.04 mM |
| GSH | — | — | 1 mM |
| DsbC | — | — | 5 µM |
| Water added | 1.925 µL | 1.625 µL | 0.250 µL |
Fixed Components
Each 20 µL reaction also contained:
- 5 µL lysate
- 2 µL DNA template
- 2 µL base buffer
sfGFP and PETase used the standard reducing lysate, while Reteplase used the designated Δgor oxidizing/DSB lysate with IAM treatment and cysteine quenching.
Design Rationale
All three formulations use the same low-cost glucose + ribose + NMP metabolic backbone so that differences in functional protein production can be interpreted more clearly.
- sfGFP: simple reducing baseline with no additional folding reagents.
- PETase: kept in the assigned reducing regime, with increased HEPES as an exploratory pH-buffering adjustment.
- Reteplase: uses GSSG/GSH + DsbC to support formation and correction of its multiple disulfide bonds.
These are first-round hypotheses rather than optimized formulations and will be refined using experimental results.
AI Use
I used ChatGPT (OpenAI, GPT-5.6 Sol) as a learning and design-support tool for this homework. I used it to:
- explain CFPS concepts that were new to me,
- compare literature reaction formulations,
- identify differences between PEP + NTP and Ribose + NMP strategies,
- connect CFPS formulation principles to my own biomanufacturing research,
- review literature beyond the papers provided in the assignment,
- brainstorm first-round reaction compositions for sfGFP, PETase, and Reteplase, and
- suggest a possible community reagent for improving disulfide-bond formation.
AI-generated ideas were treated as design hypotheses, not established results, and were checked against the assignment materials and published literature before being included.
Representative AI Prompts
Understanding reaction composition
Why do potassium glutamate and magnesium glutamate concentrations differ between CFPS formulations, and which is more sensitive?
Energy and nucleotide strategy
Explain the difference between a PEP + NTP reaction formulation and a Ribose + NMP formulation.
Literature comparison
Compare PANOx-SP from Jewett & Swartz (2004) with the RFopt formulation from Olsen et al. (2026), focusing on reagent choice, concentration, energy regeneration, nucleotide strategy, and cost.
Broader literature review
Review additional CFPS literature beyond the papers provided in the homework and identify important developments in energy regeneration, nucleotide recycling, reaction simplification, DOE, active learning, and AI-guided optimization.
Reaction design
Using the homework constraints and published CFPS literature, propose first-round 20 µL reaction designs for sfGFP, PETase, and Reteplase. Explain the reasoning for salts, energy sources, building blocks, cofactors, and redox components.
Community reagent
Suggest a reagent that could be added to the Community Reagents inventory, especially one that may improve functional folding of Reteplase, and support the suggestion with literature.
AI client: OpenAI ChatGPT Model: GPT-5.6 Sol Role: Learning companion, research/design support, and scientific communication assistance Accessed: October 2026
References
Cell-Free Protein Synthesis Formulations
Jewett MC, Swartz JR. Mimicking the Escherichia coli Cytoplasmic Environment Activates Long-Lived and Efficient Cell-Free Protein Synthesis. Biotechnology and Bioengineering. 2004;86(1):19–26.
https://doi.org/10.1002/bit.20026Calhoun KA, Swartz JR. Energizing Cell-Free Protein Synthesis with Glucose Metabolism. Biotechnology and Bioengineering. 2005;90(5):606–613.
https://doi.org/10.1002/bit.20449Olsen ML, Copeland CE, Sundberg CA, Aw R, Shaver ZM, Rao G, Swartz JR, Karim AS, Jewett MC. Design-Driven Optimization of Low-Cost Reagent Formulations for Reproducible and High-Yielding Cell-Free Gene Expression. Nature Communications. 2026;17:3478.
https://doi.org/10.1038/s41467-026-69605-8
Cell-Free Biomanufacturing and Optimization
Silverman AD, Karim AS, Jewett MC. Cell-Free Gene Expression: An Expanded Repertoire of Applications. Nature Reviews Genetics. 2020;21:151–170.
https://doi.org/10.1038/s41576-019-0186-3Rasor BJ, Vögeli B, Landwehr GM, Bogart JW, Karim AS, Jewett MC. Toward Sustainable, Cell-Free Biomanufacturing. Current Opinion in Biotechnology. 2021;69:136–144.
https://doi.org/10.1016/j.copbio.2020.12.012Borkowski O, Koch M, Zettor A, et al. Large Scale Active-Learning-Guided Exploration for In Vitro Protein Production Optimization. Nature Communications. 2020;11:1872.
https://doi.org/10.1038/s41467-020-15798-5Zhu J, Meng Y, Gao W, et al. AI-Driven High-Throughput Droplet Screening of Cell-Free Gene Expression. Nature Communications. 2025;16:2720.
https://doi.org/10.1038/s41467-025-58139-0
AI-Guided CFPS Design
- Smith AA, Wong EL, Donovan RC, et al. Using a GPT-5-Driven Autonomous Lab to Optimize the Cost and Titer of Cell-Free Protein Synthesis. bioRxiv. 2026.
https://doi.org/10.64898/2026.02.05.703998
Reteplase and Disulfide-Bond Folding
- Zhuo XF, Zhang YY, Guan YX, Yao SJ. Co-Expression of Disulfide Oxidoreductases DsbA/DsbC Markedly Enhanced Soluble and Functional Expression of Reteplase in Escherichia coli. Journal of Biotechnology. 2014;192:197–203.
https://doi.org/10.1016/j.jbiotec.2014.10.028
Software and Online Resources
HTBAA CFPS Coopetition.
Used to design and submit the 20 µL CFPS reaction compositions for sfGFP, PETase, and Reteplase.
Accessed October 2026.OpenAI ChatGPT.
GPT-5.6 Sol was used as a learning companion, literature-review assistant, and design-support tool for comparing CFPS formulations and developing first-round reaction hypotheses.
Accessed October 2026.